AFFECTIVE LEARNING ESIGN A PRINCIPLED APPROACH TO … · Affective Learning Design: A Principled...

AFFECTIVE LEARNING DESIGN:

A PRINCIPLED APPROACH TO

EMOTION IN LEARNING

Jonathon Jeffs Headrick

Master of Applied Science (research)

Bachelor of Applied Science (Human Movement Studies)

Submitted in fulfilment of the requirements for the degree of

Doctor of Philosophy

School of Exercise and Nutrition Sciences

Faculty of Health

Queensland University of Technology

October 2015

Affective Learning Design: A Principled Approach to Emotion in Learning i

Keywords

Action; Affect; Affective Constraints; Affective Learning Design; ALD;

Cognition; Constraints; Cricket; DST; Dynamical Systems; Ecological Dynamics;

Ecological Psychology; Emotion; Intention; Learning; Representative Learning

Design; RLD; Metastability; Self-organisation; SLEQ; Sport

ii Affective Learning Design: A Principled Approach to Emotion in Learning

Abstract

This PhD programme set out to explore the role of emotion during learning in

sport and provide evidence of how action, emotion and cognition might interact

under the influence of targeted manipulations to constraints. Through theoretical

modelling and applied findings, emotion has been advocated as an integral part of

learning environments given the situational information variables that influence a

learner’s engagement and approach to a task. After reviewing relevant literature in

the area, it was established that a principled approach for considering emotion in

learning was lacking, particularly in the context of applied sport research. However,

some examples of theoretical modelling were found, conceptualising individuals as

dynamic systems, incorporating the self-organising tendencies of actions, emotions,

and cognitions. Taking these conceptualisations into account the first major

contribution of this thesis is the development of a principled approach to emotion in

learning, Affective Learning Design (ALD). This concept advocates for: (i) the

design of emotion-laden learning environments and (ii) the holistic recognition of

individual emotion and coordination tendencies during learning. The term of

affective constraints was also introduced referring to the manipulation of affective

variables that have the potential to influence the emergent behaviour of learners

(Chapter 3).

Based on the principles of ALD the subsequent chapters of the thesis set out to

investigate how affective constraints could be incorporated and monitored in applied

sport contexts. The first stage of this process was to decide on an appropriate

measurement tool to adequately track emotion across specific time scales. A review

of the existing methods revealed a dearth of appropriate tools and therefore the

Affective Learning Design: A Principled Approach to Emotion in Learning iii

development of a new method was warranted. Presented in Chapter 4, the

development of the Sport Learning and Emotions Questionnaire (SLEQ) is an

important output of this PhD given its specificity to learning environments in sport.

The SLEQ provides an indication of emotion intensity overall, and in respect to four

distinct subscales (Enjoyment, Nervousness, Fulfilment, and Anger). The

questionnaire can be implemented at several time points to track fluctuations in

emotion that are indicative of individualised tendencies following task

manipulations. This tool was developed within the context of learning in sport, for

use during learning in sport, and therefore provides a new method of analysis with

implications for researchers and practitioners alike.

In order to demonstrate the effectiveness of ALD and the SLEQ in practice, the

next stage of the PhD explored the interaction between actions, emotions, and

cognitions in applied sport environments. Chapter 5 set about observing emotion

intensity alongside action in systematically constrained games of the passing and

possession game ‘Endball’. Individual possession time was manipulated across four

4 v 4 games with the aim of producing observable changes in emotions and

performance characteristics. Through the analysis of SLEQ and game event data

(e.g. complete passes, goals, errors) clear interactions between SLEQ scores (e.g.

Enjoyment subscale) and action (e.g. goals, complete passes) were observed,

particularly during the transition from one game to the next. The second applied case

example in Chapter 5 took this approach further, incorporating a measure of

cognition in the form of confrontational interviews in a cricket batting task. In this

case a more in-depth individualised approach was adopted providing detailed

measures of movement characteristics, performance outcomes, emotion intensity,

intentions, and game plans. To manipulate the demands of the task the distance of

iv Affective Learning Design: A Principled Approach to Emotion in Learning

the pitch was shortened to replicate deliveries of increased speeds, by the same

bowler. By collectively analysing all categories of variables, the critical links

between actions, emotions, and cognitions were able to be observed across several

time points. Findings revealed that shorter foot movement distances were associated

with increased enjoyment, perception of achievement, and runs scored, particularly

when the simulated delivery speed was increased from 125km/h – to – 130km/h.

Therefore, both of these applied studies have demonstrated the novel approach

advocated by the concept of ALD, highlighting how affective constraints can be

incorporated in learning tasks, and the importance of considering actions, emotions,

and cognitions in unison.

Chapter 6 of the thesis proposed a model of Affective Learning Design that

draws on the theoretical conceptualisations, and highlights the findings of the two

applied studies. This model advocates for ALD principles to be applied over

interacting time scales, with an emphasis on individualised study and/or practice

designs. Each of the four model phases (Evaluation, Planning, Implementation, and

Observation / Monitoring) are informed by the experiential knowledge of a coach or

practitioner, along with theoretical underpinnings, such as those discussed

throughout this PhD programme. Through the discussion of these phases in relation

to the two examples from Chapter 5, the practicality and relevance of this model for

future applied work in sport is highlighted.

Summarising the theoretical and practical implications of this thesis highlights

the major contribution of this PhD programme to the fields of skill development,

motor learning, and applied sport psychology. The theoretical conceptualisations of

ALD and affective constraints provide a pivotal framework that can be incorporated

into future discussions regarding the crucial role of emotion in learning.

Affective Learning Design: A Principled Approach to Emotion in Learning v

Furthermore, these conceptualisations also inform the design of future sport learning

tasks, advocating strongly for the recognition and consideration of the intertwined

relationships between actions, emotions, and cognitions. The practical component of

this thesis developed a new emotion questionnaire (SLEQ), targeted specifically at

tracking emotion intensity throughout learning tasks in sport. To exemplify the

application of the SLEQ and ALD principles, two studies were designed to adopt

these innovative approaches in both group and individualised examples. Finally, a

model of ALD was conceptualised combining the key ideas and findings of the PhD

into a succinct and accessible format with various implications available to coaches,

researchers and practitioners.

Together the insightful theoretical conceptualisations, innovative

methodological developments, rich findings, and abundance of practical implications

demonstrate why the tangible outputs of this PhD are so critical to enhancing of

learning environments in sport. Emotion must therefore be recognised as a key

consideration in the design of representative learning tasks, alongside the actions,

and cognitions of an individual.

vi Affective Learning Design: A Principled Approach to Emotion in Learning

Table of contents

Keywords .................................................................................................................................. i

Abstract .................................................................................................................................... ii

Table of contents ..................................................................................................................... vi

List of figures .......................................................................................................................... ix

List of tables ............................................................................................................................ xi

List of abbreviations .............................................................................................................. xiii

Statement of original authorship ............................................................................................ xv

Acknowledgements ............................................................................................................... xvi

Research outputs ................................................................................................................. xviii

Chapter 1: Introduction............................................................................................. 1

Thesis structure ...................................................................................................................... 10

Chapter 2: Literature review .................................................................................. 13

An ecological dynamics approach .......................................................................................... 13

A working definition of emotion ............................................................................................ 20

Emotions and human behaviour ............................................................................................. 21

Intentionality .......................................................................................................................... 28

Representative design ............................................................................................................. 29

Representative learning design ............................................................................................... 30

Learning and performance ...................................................................................................... 32

Goal orientation ...................................................................................................................... 34

Motivation .............................................................................................................................. 37

Summary ................................................................................................................................ 39

Chapter 3: Affective learning design: Developing a principled approach to

emotion in learning ................................................................................................... 41

Abstract .................................................................................................................................. 42

Introduction ............................................................................................................................ 43

Affective learning design ....................................................................................................... 50

Affective learning design in practice ...................................................................................... 55 The individualisation of affect ................................................................................................... 56 Time-scales and affects .............................................................................................................. 58 Emotions are embedded in situation-specific task constraints ................................................... 59

Conclusions ............................................................................................................................ 64

Chapter 4: Development of a tool for monitoring emotions during learning in

sport ............................................................................................................... 67

Overview ................................................................................................................................ 68

Introduction ............................................................................................................................ 69

Affective Learning Design: A Principled Approach to Emotion in Learning vii

Development of the Sport Learning and Emotions Questionnaire (SLEQ) ............................73

Chapter 4: Phase 1 – Identifying emotional items ................................................ 75

Methods ..................................................................................................................................75 Participants ................................................................................................................................ 75 Procedure ................................................................................................................................ 76 Analysis ................................................................................................................................ 76

Results & discussion ...............................................................................................................76

Chapter 4: Phase 2 - Face validation of items ....................................................... 83


Results & discussion ...............................................................................................................84

Chapter 4: Phase 3 – Assessment of item and factor structure ........................... 89


Results & discussion ...............................................................................................................94 Descriptive results ...................................................................................................................... 94 Exploratory factor analysis ........................................................................................................ 96 Confirmatory factor analysis .................................................................................................... 101 Sport learning and emotion questionnaire design .................................................................... 108

Chapter 5: Implementing the SLEQ during learning in sport: Applied case

examples ............................................................................................................ 115

Chapter 5: Applied case example A ..................................................................... 119

Abstract .................................................................................................................................120

Introduction ...........................................................................................................................121

Methods ................................................................................................................................124 Participants .............................................................................................................................. 124 Procedure .............................................................................................................................. 124 Analysis .............................................................................................................................. 127

Results ...................................................................................................................................128

Discussion .............................................................................................................................141 Between team findings ............................................................................................................. 141 Overall correlation findings ..................................................................................................... 142 Implications ............................................................................................................................. 149 Limitations and future research ................................................................................................ 150 Conclusion .............................................................................................................................. 151

Chapter 5: Applied case example B ...................................................................... 153

Abstract .................................................................................................................................154

Introduction ...........................................................................................................................155

Methods ................................................................................................................................159 Participant .............................................................................................................................. 159 Procedure .............................................................................................................................. 159 Analysis .............................................................................................................................. 164

viii Affective Learning Design: A Principled Approach to Emotion in Learning

Results .................................................................................................................................. 166 Over comparisons .................................................................................................................... 166 Ball number comparisons ........................................................................................................ 167 Change in z score correlations ................................................................................................. 167 Question and answer responses ............................................................................................... 168 Over by over summary ............................................................................................................ 172

Discussion ............................................................................................................................ 178 Action findings ........................................................................................................................ 178 Emotion findings ..................................................................................................................... 180 Cognition findings ................................................................................................................... 181 Combined findings ................................................................................................................... 182 Over by over summaries .......................................................................................................... 184 Implications ............................................................................................................................. 187 Limitations and future research ............................................................................................... 188 Conclusion 189

Chapter 6: A conceptual model of affective learning design .............................. 191

Abstract ................................................................................................................................ 192

Introduction .......................................................................................................................... 193 Why it is important to consider emotion in learning ................................................................ 193 The need for a principled integration of emotion into learning ............................................... 196

An ecological dynamics approach to emotion and learning ................................................. 197 Ecological psychology ............................................................................................................. 198 Dynamic systems theory and emotions .................................................................................... 200 Learning ............................................................................................................................. 202

A model of affective learning design ................................................................................... 204 Evaluation ............................................................................................................................. 205 Planning ............................................................................................................................. 207 Implementation ........................................................................................................................ 209 Observation and monitoring .................................................................................................... 211

Examples of ALD principles in practice .............................................................................. 214 Example 1: Action-emotion relationships in a team passing game.......................................... 214 Example 2: Action, emotion, and cognitions in a representative cricket batting task .............. 215 Summary ............................................................................................................................. 217

Chapter 7: Epilogue ............................................................................................... 219

Introduction .......................................................................................................................... 220

Theoretical findings and implications .................................................................................. 220

Practical findings and implications ...................................................................................... 224

Limitations and future research ............................................................................................ 231

Conclusion ............................................................................................................................ 233

Bibliography ........................................................................................................... 235

Appendices ............................................................................................................. 257 Appendix A – Selected interpretations of emotion .................................................................. 257 Appendix B – Emotions during learning in sport survey (phase 1) ......................................... 261 Appendix C – A Preliminary exploration of emotion items in a golf putting task .................. 263 Appendix D – Emotions during learning in sport survey (phase 2) ......................................... 287 Appendix E – Emotions during learning in sport survey (phase 3) ......................................... 290 Appendix F – Perception of speed scale .................................................................................. 293 Appendix G – Confrontational interview questions and answers ............................................ 295

Affective Learning Design: A Principled Approach to Emotion in Learning ix

List of figures

Figure 1.1. Thesis overview and structure .............................................................. 12

Figure 4.1. Scree plot displaying Eigenvalues for phase 1 items.

Components to the left of the dashed vertical line were retained

(1-39). .................................................................................................. 80

Figure 4.2. Model 1 with modifications. Values on model represent (left-

to-right) error covariances, standardised factor loadings, and

factor correlations. ............................................................................. 103

Figure 4.3. Model 2 with modifications. Values represent (left-to-right)

error covariances, standardised factor loadings, and factor

correlations. ....................................................................................... 105

Figure 5.1. Representation of session design with SLEQ and perception of

difficulty (PoD) collection occasions. ............................................... 125

Figure 5.2. Mean team SLEQ scores plotted with: (a) goals scored; (b)

touches; and (c) errors across the session. *Significant

differences in SLEQ scores between teams (p < .05) ....................... 131

Figure 5.4. Mean (combined 1st and 2

nd) foot movement, total runs, and

SLEQ score throughout the 10 over session. * significant

difference in foot movement (p <.05) ............................................... 168

Figure 5.5. SLEQ subscale scores during the session and PANAS scores

Pre and post (over 10) session. .......................................................... 169

Figure 5.6. Ball-by-ball runs scored and SLEQ score at the end of each over ..... 170

Figure 5.7. Set 1 (over 1 – 2) summary ................................................................. 173



Figure 5.10. Set 4 (over 7 – 8) summary ............................................................... 176

Figure 5.11. Set 5 (over 9 – 10) summary ............................................................. 177

Figure 6.1. Changes in complex system stability. (1) represents a weak

attractor state within a shallow (unstable) well. (2) represents a

metastable region where the system lingers between attractor

states (i.e. 1 & 3). (3) represents a strong attractor state with a

deep (stable) well .............................................................................. 201

Figure 6.2. A model of Affective Learning Design .............................................. 204

Figure A1. Representation of trends between emotion items and putting

scores across sets. .............................................................................. 274

Figure A2. Correlations of changes in z scores for putting performance and

number of positive item selections between sets. .............................. 277


number of negative item selections between sets. ............................. 278

x Affective Learning Design: A Principled Approach to Emotion in Learning


number of (a) positive and (b) negative item selections across

the session(Sets 1-5) .......................................................................... 280

Affective Learning Design: A Principled Approach to Emotion in Learning xi

List of tables

Table 4.1. The list of 39 most identified emotional terms from the phase 1

survey. Items are split into those that are common with the

preliminary 39 item SEQ (Jones et al., 2005), and those unique

to this study. ........................................................................................ 79

Table 4.2. Percentage of participants reporting items to be relevant to

learning a skill in sport, and to any of the five key emotions.

Items are presented in descending order of overall relevance............. 86

Table 4.3. Summary of survey results presented in descending order of

mean rating. ......................................................................................... 95

Table 4.4. Item factor loadings, eigenvalues, variance %, and Cronbach’s α

for each of the four extracted factors and/or questionnaire

subscales. ............................................................................................. 97

Table 4. 5. Proposed factor and item structure following EFA. Item origin:

Phase 1: item originates from phase 1 list only; Both: item

originates from the preliminary SEQ and Phase 1. ........................... 100

Table 4.6. Factor loadings, error variances, and subscale reliability (α) for

model 2 following CFA. Item origin: Phase 1: item originates

from phase 1 list only; Both: item originates from the

preliminary SEQ and Phase 1. ........................................................... 107

Table 4.7. Mean score and correlations for each factor / subscale. ....................... 107

Table 5.1. Summary of game events, mean perception of difficulty (PoD)

ratings, and mean SLEQ scores for each of the two teams. .............. 130

Table 5.2. Change in z score correlations between observed variables from

Pre to Game 1 – 1st half (SLEQ only). .............................................. 132


Pre to Game 4 – 2nd half (SLEQ only). ............................................ 132


Game 1 – 1st half to Game 1 – 2nd half (3 second game). ............... 133


Game 1 – 2nd half to Game 2 – 1st half (3 second game – 2

second game). .................................................................................... 134


Game 2 – 1st half to Game 2 – 2nd half (2 second game) .............. 135



second game) ..................................................................................... 136



xii Affective Learning Design: A Principled Approach to Emotion in Learning



second game) ..................................................................................... 138




Game 1 – 1st half to Game 4 – 2nd half (whole session) ................ 140

Table 5.11. Delivery speed calculations ................................................................ 161

Table 5.12. Change in z score correlations for the session. * Significant

correlation (p < .05) ........................................................................... 171

Table A1. Summary of putting scores and emotion items for each set. ................ 269

Table A2. Count of emotion items selected for each set and respective totals

presented in descending order. Origin column indicates where

the item originated: SEQ - exclusively from the preliminary list

of 39 items of Jones et al. (2005); Phase 1 - exclusively from

the list of unique items in phase 1; Both – from both the

preliminary SEQ and phase 1. ........................................................... 271

Table A3. Pearson (r) correlation coefficients [95% confidence intervals]

for changes in z scores between putting performance and

emotion item selection. S – putting score; P – positive items; N

– negative items. No relationships were statistically significant

at the .05 level.................................................................................... 275

Table A4. Items grouping to the three clusters and the origin of these items

(Phase 1 list or preliminary SEQ). Mean values display the

predominance of Cluster 3 items for all occasions where

emotion items were selected. Significant differences between

clusters indicated for: cluster 1 and 2 a ; cluster 2 and 3

b ;

cluster 1 and 3 c. ................................................................................ 282

Table A5. Comparison of participant gender breakdown, putting scores (S),

positive items (P), and negative items (N) between the two

clusters for the five putting sets. *Significant differences at the

.05 level. ............................................................................................ 284

Affective Learning Design: A Principled Approach to Emotion in Learning xiii

List of abbreviations

ALD – Affective Learning Design

ANG – Anger subscale of SLEQ

CFA – Confirmatory Factor Analysis

CFI – Comparative Fit Index

DS – Dynamic Systems

DST – Dynamic Systems Theory

EFA – Exploratory Factor Analysis

ENJ – Enjoyment subscale of SLEQ

FoBS – Forcefulness of Bat Swing

FUL – Fulfilment subscale of SLEQ

KMO – Kaiser-Meyer-Olkin measure of sampling adequacy

MI – Modification Indices

NERV – Nervousness subscale of SLEQ

PoA – Perception of Achievement rating scale

PoD – Perception of Difficulty rating scale

PoS – Perception of Speed rating scale

QoC – Quality of Contact

RLD – Representative Learning Design

RMSEA – Root Mean Square Error Approximation

xiv Affective Learning Design: A Principled Approach to Emotion in Learning

SEM – Structural Equation Modelling

SEQ – Sport Emotion Questionnaire

SLEQ – Sport Learning and Emotions Questionnaire

χ2 – Chi-Square

Affective Learning Design: A Principled Approach to Emotion in Learning xv

Statement of original authorship

The work contained in this thesis has not been previously submitted to meet

requirements for an award at this or any other higher education institution. To the

best of my knowledge and belief, the thesis contains no material previously

published or written by another person except where due reference is made.

Signature:

Date: _____28th October 2015 _

QUT Verified Signature

xvi Affective Learning Design: A Principled Approach to Emotion in Learning

Acknowledgements

I have many people to acknowledge and thank for their support and

contributions throughout the PhD programme. Thank you to:

- QUT, the school of ENS, Faculty of Health, and IHBI for their support in

terms of funding and resources, without which the PhD would not have

been as successful.

- All participants who agreed to be involved in each stage of the PhD project.

- ENS staff and fellow post grad students who assisted with data collection,

and survey distribution. Particular thanks to Steve Duhig, Lee Wharton,

Michael Cook, Brendan Moy, and Geoff Minett for their contributions in

support of the project.

- Fellow ENS post grad students and staff members past and present who

have shared the many fun times, and helped me through the rough times.

There are too many of you to mention, but you know who you are.

- Other students supervised by Ian and/or Keith who have helped me out or

made me laugh including: Elissa Phillips, Sian Barris, Matt Dicks, Dan

Greenwood, Mike Maloney, Chris McCosker, and Jono Connor.

- Thesis examiners and seminar panel members for their valuable input and

support shown towards the project.

- Duarte Araújo and Tony Oldham for their guidance, support, and expertise

through various phases of the PhD programme, despite not officially being

on the supervisory ‘team’.

QUT Verified Signature

Affective Learning Design: A Principled Approach to Emotion in Learning xvii

- Ross Pinder for allowing me to incorporate many of the ideas from his PhD

and research into this thesis, his role as an unofficial supervisor, and most

importantly being a good mate (Thanks Boss).

- Keith Davids for his continual guidance and encouragement in my

development as a researcher whether in the form of an early morning email

from the other side of the world, a page full of suggested changes, or a

sledgehammer like response to reviewer comments on my behalf.

- Special thanks to Ian Renshaw for his tireless advice, guidance,

encouragement, and entertainment provided throughout the course of my

post grad studies. I can safely say that my time as a PhD student would not

have been so rewarding and enjoyable without your enthusiasm and

dedication towards my work, your catalogue of cricket stories, and the

many coffees that I owe you.

- Last but not least, huge thanks and love to Mum, Dad, Mitch, Kara, along

with my extended family, and friends for their understanding (well sort of)

and support of my ‘scholarly’ activities throughout.

xviii Affective Learning Design: A Principled Approach to Emotion in Learning

Research outputs

Peer reviewed journal articles:

Headrick, J., Renshaw, I., Davids, K., Pinder, R. A., & Araújo, D. (2015). The

dynamics of expertise acquisition in sport: The role of affective learning

design. Psychology of Sport and Exercise, 16, 83-90. doi:

10.1016/j.psychsport.2014.08.006

Articles in preparation for submission:

Headrick, J., Renshaw, I., Davids, K., Pinder, R. A., & Araújo, D. (in preparation). A

conceptual model of affective learning design. International Review of Sport

and Exercise Psychology.

Headrick, J., Renshaw, I., Oldham, A. R. H., Davids, K., & Pinder, R. A. (in

preparation). Development of the Sport Learning and Emotions Questionnaire

(SLEQ). Journal of Sport & Exercise Psychology.

Book chapters:

Pinder, R. A., Headrick, J., & Oudejans, R. R. D. (2015). Issues and challenges in

developing representative tasks in sport. In D. Farrow & J. Baker (Eds.), The

Routledge Handbook of Sports Expertise (pp. 269-281). London: Routledge.

Pinder, R. A., Renshaw, I., Headrick, J., & Davids, K. (2014). Skill acquisition and

representative task design. In K. Davids, R. Hristovski, D. Araújo, N.

Balagué - Serre, C. Button & P. Passos (Eds.), Complex Systems in Sport (pp.

319-333). London: Routledge.

Affective Learning Design: A Principled Approach to Emotion in Learning xix

Conference papers:

Renshaw, I., Headrick, J., & Davids, K. (2014). Affective learning design: Building

emotions into representative learning design. Paper presented at the

International Conference on Complex Systems and Applications, Le Havre,

France.

Conference presentations (oral):

Headrick, J. (2015). Affective learning design and the role of emotions during

learning in sport. Paper presented at the Australasian Skill Acquisition

Research Group Conference – 12-14 June, Perth, Australia.

Conference presentations (poster)

Headrick, J., Renshaw, I., Davids, K. (2014). A conceptual model of affective

learning design. Institute of Health and Biomedical Innovation conference –

20-21 November, Gold Coast, Australia.

Invited presentations:

Headrick, J. (2015). The role of emotion during learning in sport. Queensland

Academy of Sport – Performance Science Unit – 14 January, Brisbane. Australia.

Other:

Headrick, J. (2015). Affective learning design: A principled approach to emotion in

Learning. PhD Final Seminar – 29 May, Queensland University of

Technology, Brisbane, Australia.

Headrick, J. (2013). A principled approach to emotion in Learning. PhD

Confirmation Seminar – 13 March, Queensland University of Technology,

Brisbane, Australia.

xx Affective Learning Design: A Principled Approach to Emotion in Learning

You must always believe you will become the best, but you must never

believe you have done so

Juan Manuel Fangio

Chapter 1: Introduction 1

Chapter 1: Introduction

This chapter introduces the scope of the PhD programme and identifies the key

theoretical and experimental shortcomings of the literature that will be discussed and

examined throughout the following sections. Towards the end of the chapter an

overview of the thesis structure is presented to highlight how each of the chapters fits

within the thesis.

Ideas and concepts presented in this introductory chapter have also been

incorporated into the following peer reviewed research outputs:


developing representative tasks in sport. In D. Farrow & J. Baker (Eds.), The




Balagué - Serre, C. Button & P. Passos (Eds.), Complex Systems in Sport (pp.





France.

2 Chapter 1: Introduction

Introduction

“Life is essentially a process of dynamic reorganisation, and therefore

emotions are an inevitable part of the life process itself”

(Jarvilehto, 2000a, p. 56)

Learning is an inherently emotional experience where an individual is

frequently exposed to periods of success, failure and challenges from both physical

and psychological perspectives (Davids, 2012; Seifert, Button, & Davids, 2013).

Each individual arrives at a learning experience with specific capabilities that must

be modified or adapted in order to meet the demands of the new task (Kelso, 2003).

Therefore a learning experience must be considered in relation to the individualised

approach to concepts including perception, intentions, attention, cognitions, and

emotions (Davids, 2012; Kelso, 2003).

Tasks that are emotion-laden are considered to facilitate a ‘deeper’ engagement

for learning and performance (Jones, 2003; Solomon, 2008). Indeed, emotional

engagement is seen as being essential for effective learning (Pessoa, 2011).

However, the role of emotion during learning has often been neglected because

emotion-laden responses are considered irrational or instinctive, and therefore

perceived as negative (Hutto, 2012; Jarvilehto, 2000a; Lepper, 1994). Furthermore,

the influence of emotion has historically been portrayed as a disturbance to the

acquisition of knowledge or expertise (Jarvilehto, 2000a). Emotionless responses

made from a purely informational stance have been described as ‘cold cognition’,

whereas emotion-laden responses are viewed as ‘hot cognition’ (Abelson, 1963;

Lepper, 1994). The expression of ‘sit on your hands’ in relation to choosing a move

in a game of chess is an example of the view that it is necessary to suppress or

remove emotions in order to make rational decisions (i.e. cold cognition) (Charness,


Tuffiash, & Jastrzembski, 2004). Crucially in sporting contexts learners are often not

afforded this ‘thinking’ time and must therefore act instinctively based on the

interaction between their perceptions of the task and pre-existing physical, cognitive,

and emotional capabilities (Davids, 2012). Therefore there is a need for an accurate

and detailed description of emotional experiences in sport as emotions are often

under-estimated or ignored, perhaps due to the paucity of a theoretical framework in

sporting contexts (Hanin, 2007b; Vallerand & Blanchard, 2000). Additionally,

progress in understanding emotions has also been limited by traditional linear

cognitive thinking perpetuating the debate over the pre-eminence of cognition over

emotion; where cognitions of events are thought to result in emotional reactions

based on ‘inner’ processes or knowledge (Jarvilehto, 1998a, 2000a, 2009; Lewis &

Granic, 2000). This outdated reductionist approach (discussed further in later

sections) to understanding emotions by cognitivists has hampered the ability to

model relations between goals, emotions and emotion regulation (Kiverstein &

Miller, 2015; Lewis & Granic, 2000). However, some psychologists have recently

begun to acknowledge the advantages of nonlinear dynamic systems (DS) reasoning

in explaining behaviour and this has led to the emergence of a DS perspective of

emotional development (Jarvilehto, 2000a, 2001; Lewis, 1996; Lewis & Granic,

2000). Yet to be seen in the literature, however, is a principled exploration of the role

of emotions in learning for sports performance.

Dynamic systems approach

Complex dynamic systems, such as humans, have the capability to self-

organise their actions or behaviour to achieve specific task objectives without direct

input from higher order structures or predetermined rules (Kelso, 1995; Lewis,

2000b). Self-organising processes take place under the influence of organismic


(individual), environmental and task constraints (see Newell, 1986) of the specific

performance environment. Interacting constraints (individual – environment) can

both limit and guide behaviour, shaping the affordances which are perceived to

complete goal-directed action (Gibson, 1979; Newell, 1986). Through the detection

of informational variables and subsequent perception of possible affordances, stable

patterns of behaviour emerge; referred to as attractor states. Fluctuations in

information flows (through changes in constraints) have the potential to perturb

stable (attractor) states of behaviour and facilitate phase transitions to new states,

requiring the system to adopt different states of organisation (Kelso, 1995, 2012;

Kelso & Tognoli, 2009). By adopting novel and functional patterns of behaviour a

complex system (i.e. performer) is considered to have gone through a process of

learning or development, whereby stable patterns of behaviour become characteristic

responses to specific information flows between the performer and environment

(Lewis, 2000b; Thelen, 2002; Thelen & Smith, 1994). During learning the state of

the system becomes unstable as learners search for functional coordination solutions

(attractors) to the new constraints of the environment (Chow et al., 2007). Markers

or predictors for phase transitions include increased variability in movements as

learners are forced into metastable regions of performance (Chow, Davids,

Hristovski, Araújo, & Passos, 2011; de Weerth & van Gert, 2000) From this

complex systems approach, the role of the affective domain has yet to be fully

incorporated into the study of learning, in regards to understanding how emotions

can be linked with emergent human behaviour in contexts such as sport.

Affect, behaviour, and cognition

Emotion will be the focus of this thesis, but it must be acknowledged that the

term emotion is part of the larger group of affective phenomena. The term ‘Affect’


incorporates phenomena that occur over different time scales such as emotions,

feelings, moods and personality traits (Lewis, 2000a; Vallerand & Blanchard, 2000).

Emotion can be distinguished from the concepts of mood and feeling, by a stronger

affective state, sudden onset, and a relationship with an object, event or person (Moll,

Jordet, & Pepping, 2010; Oatley & Jenkins, 1996; Zadra & Clore, 2011). Feelings

relate more directly to subjective experiences emerging from a task, without

physiological or behavioural changes. Moods do not display a relationship with an

object, event or person, develop over longer time scales, and often emerge from an

initial emotion (Frijda, 1994; Vallerand & Blanchard, 2000). Traits are more

permanent and stable affective tendencies that have developed into inherent aspects

of an individual’s personality (Watson & Clark, 1994).

Affect, along with behaviour and cognition form a triad that represent feeling,

acting, and knowing respectively (Breckler, 1984; Hilgard, 1980). Behaviour

encompasses verbalisations about acting, intentions to act, and overt actions which in

their simplest form involve moving towards or away from an object. Cognition

frames our knowledge or intentions towards an object including thoughts, beliefs and

perceptual reactions (Ajzen, 1989; Breckler, 1984; Hilgard, 1980). Similar

conceptualisations are also captured by Bloom’s (1956) Taxonomy, which described

the cognitive, affective, and psychomotor domains within education.

The interaction between affect, behaviour and cognition can be described from

a complex (dynamic) systems approach where the three components influence each

other to form emergent appraisals of experiences (Frijda, 1993; Lewis, 1996). A

complex systems approach considers an individual as many interacting parts that

self-organise under the influence of constraints (see Newell, 1986) to achieve

specific objectives (Kelso, 1995). For example, an individual experiencing an


emotional event must perceive how an event or object relates to their own intrinsic

dynamics to form emergent behavioural (action), cognitive, and affective responses

(Bower, 1981; Ortony, Clore, & Collins, 1994). In this case, knowing of an

object/event constrains a functional emotional response, as the same information may

be construed differently by individuals (Jarvilehto, 2000a). That is, an object/event

affording emotional behaviours. Similarly, the behaviour of an individual can be

influenced by knowledge of an object and the emotional attachment that has

previously been construed, such as a challenging rock climbing hold from which the

climber previously fell or slipped (Davids, Brymer, Seifert, & Orth, 2014; Kiverstein

& Miller, 2015; Lewis, 1996). Emotion then can be considered as the ‘readiness to

act’ in an intentional manner (Frijda, 1986; Jarvilehto, 2001). Therefore, in the

regulation of human behaviour the perceptions, actions and cognitions of an

individual self-organize to form emergent physical and psychological responses

(Warren, 2006).

Cognition and emotion

The relationship between cognition and emotion in particular has been

discussed extensively from a complex systems perspective by Lewis (Lewis, 1996,

2000a, 2002, 2004). From this perspective emotions shape cognitions and these

cognitions further influence emotions to form stable attractor states of system

organisation (Kelso, 1995; Lewis, 2004). The cognition-emotion link must also be

considered over interacting time scales, which influence each other in a reciprocal

relationship to form characteristic personality traits, or shape immediate emotional

responses (Lewis, 2002; Newell, Liu, & Mayer-Kress, 2001). Furthermore, the

influential relationship between cognition and emotion has been found to distinguish

the intensity at which individuals experience emotions. In this case the


individualised nature of cognitions and thoughts were linked to the intensity of both

positive and negative emotions when individuals were exposed to emotion inducing

information or stimuli (Larsen & Diener, 1987; Larsen, Diener, & Cropanzano,

1987). More specifically, individuals displaying high emotion intensities were also

found to engage in more personal and empathetic cognitive tendencies. Considering

all the above concepts in relation to learning to perform skills, a combination of

environmental (e.g. physical and visual) and individual (e.g. intentions, emotions,

motivations) information sources constrain the emergent behaviour of each

individual (Kelso, 1995; Masters, Poolton, Maxwell, & Raab, 2008; Renshaw,

Oldham, & Bawden, 2012; Zadra & Clore, 2011). Maintaining an emotional

investment or engagement in a learning task can therefore be the result of implicit

(e.g. personal goals) and/or explicit (e.g. parent, teacher, coach) influences or

challenges (Collins & MacNamara, 2012). This reinforces the importance of the

relationship between an individual and his/her learning environment.

Individual – environment relationship

A key idea in Ecological Psychology concerns the link between the perceptual

information provided by the environment and the resultant actions or behaviours of

complex systems, such as humans (Araújo, Davids, & Hristovski, 2006). In the

study of visual perception, Gibson (1979) determined that the movements of an

individual bring about changes in information flow from which affordances

(opportunities for action) are perceived to support further movement. Therefore,

information and movement are deemed to be dependent on each other in a

dynamically coupled relationship. As a result, a cyclic process is created where

action and perception symbiotically support goal-directed movement behaviour

(Davids, Button, & Bennett, 2008). Gibson (1979, p. 223) summarised this


relationship by stating that “we must perceive in order to move, but we must also

move in order to perceive”. As such, perception and action are complementary

aspects of behaviour shared between the individual and the environment, not distinct

functions separated into sensory and motor components (Jarvilehto, 1999, 2001;

Turvey, 2009). This concept provides a foundation for describing how goal directed

movement emerges throughout a wide array of applications in the study of human

behaviour. Consequently, perception-action coupling is an important consideration

for the design of skill acquisition practice tasks and investigative methodologies in

experimental research.

The work of Brunswik (1955, 1956) advocated the study of individual-

environment relations in contexts where perceptual variables that would naturally be

available to an individual are preserved or maintained (Dhami, Hertwig, & Hoffrage,

2004). To capture the concept of sampling perceptual variables from an individual’s

‘natural’ environment in an experimental task, the term representative design was

introduced (Brunswik, 1956). Therefore by incorporating representative design, the

crucial cyclical relationship between perception and action in a performance

environment is maintained (Gibson, 1979). More recent work has discussed the

concept of representative design in relation to the study of human performance and

behaviour in contexts such as sport (Araújo, et al., 2006; Araújo, Davids, & Passos,

2007). As a result, the term representative learning design (RLD) has been adopted

to highlight the importance of creating representative environments for learning and

practicing skills whereby key perceptual variables are sampled from the performance

setting to guide and restrict behaviour in the learning/practice setting (Pinder,

Davids, Renshaw, & Araújo, 2011b).


To this point, empirical work discussing RLD has focussed on visual

information provided in performance tasks (Pinder, Davids, Renshaw, & Araújo,

2011a), practice in differing environments (Barris, Davids, & Farrow, 2013), and

changes to the complexity of tasks (Travassos, Duarte, Vilar, Davids, & Araújo,

2012). However, to date no work has discussed the role of individual informational

constraints such as affect in enhancing representative learning designs (for initial

discussions see, Pinder, Renshaw, Headrick, & Davids, 2014). Some related

conceptualisations have suggested that environments may be laden with both

physical and emotional affordances that are of value to an individual (Heft, 2010;

Roe & Aspinall, 2011) The aim of this proposed programme of work is to further

consider the role of affect in learning environments with the aim of enhancing RLD.

Previous work investigating emotions in relation to human behaviour has

focussed on a ‘snapshot’ of performance at one point in time (e.g. Lane, Beedie,

Jones, Uphill, & Devenport, 2012). In contrast, the approach that will be advocated

through this current project is that learning can be best understood through adoption

of nonlinear dynamic processes, over interacting time scales. As such, the role that

emotions play in learning events is expected to be dynamic, as individuals explore

their perceptual-motor workspaces. Therefore, emotions should be studied

throughout learning periods to understand their interactions with intentions,

perceptions and actions. Furthermore, it is advocated that the presence of emotions

during learning should be embraced rather than removed or ignored in order to create

engaging and representative learning/practice environments (Renshaw, et al., 2012).

This rejects the approach taken in much of the previous work on emotions in sport

performance where emotions are seen as being either positive (e.g. joy) or negative

(e.g. fear) with respect to performance (see, Campo, Mellalieu, Ferrand, Martinent,


& Rosnet, 2012; Lane, et al., 2012; Nesse & Ellsworth, 2009). Typically,

researchers and practitioners have focussed on the negative impact of emotions;

hence viewing them as dysfunctional for the performance of a given task, or as

unwanted ‘noise’ (Davids, Glazier, Araújo, & Bartlett, 2003; Jones, 2003).

Conceptually this is similar to the way movement variability has been traditionally

viewed; it is the purpose of this PhD programme to discuss and demonstrate the need

to develop a principled approach to considering affect as a functional aspect of

learning design.

Thesis structure

This thesis is presented in a traditional format with a range of theoretical and

experimental chapters arranged to reflect the emergence or flow of ideas throughout

the PhD programme. Each chapter has been written to stand alone while also linking

with the previous and following chapters to maintain the flow of ideas and findings.

As a result, in parts there is a degree of repetition. Chapters that have been published

as journal articles, or are in the final stages of preparation for submission (3 and 6

respectively), somewhat overlap in terms of theoretical background and implications,

but have been edited to fit with the format of the thesis. Stage 1 represents the

majority of the theoretical background and literature review (Chapters 1 & 2),

culminating with Chapter 3 which is based on a published position paper. Stage 2

discusses the development of a new questionnaire for tracking emotions during

learning in sport. As such, Chapter 4 is organised into three sub-phases reflecting the

distinct methods and findings of each phase leading to the eventual questionnaire

design. Stage 3 incorporates the theoretical principles and questionnaire of the

previous stages in two preliminary experimental studies discussed in Chapter 5.

These studies demonstrate the approach advocated throughout the thesis and the


value of the newly developed questionnaire. Stage 4 brings the conceptualisations

and findings of the PhD programme together to propose a conceptual model (Chapter

6), followed by an epilogue (Chapter 7). Therefore this final stage provides

implications, limitations, and future directions of the thesis and highlights the

extensive theoretical and applied contributions of this PhD programme.


Figure 1.1. Thesis overview and structure

Chapter 2: Literature review 13

Chapter 2: Literature review

An ecological dynamics approach

Together, the theories of dynamic systems and ecological psychology are

captured by the concept of ecological dynamics (Araújo, et al., 2006; Davids &

Araújo, 2010). An ecological dynamics approach recognises the mutual relationship

between an individual and the environment to understand behaviour from an

ecological perspective with reference to dynamic systems concepts (Araújo, et al.,

2006). In comparison, traditional approaches have focussed primarily on the

processes and structures within organisms to understand behaviour, described as an

organismic asymmetry (Bentley, 1941; Davids & Araújo, 2010; Dunwoody, 2006).

Describing behaviour from the organism-environment scale takes into account how

perceptions, actions, intentions and cognitions emerge under the constraints of

information shared between the organism and environment (Seifert & Davids, 2012;

Shaw & Turvey, 1999; Warren, 2006). An ecological dynamics approach to emotion

in learning draws several comparisons with the organism-environment theory

developed in the realm of experimental psychology and mental function (Jarvilehto,

1998a, 1998b, 1999, 2000b, 2009). The organism-environment theory posits that

traditional ‘functions’ (e.g. emotion, perception, action) that contribute to the system

as a whole are not found solely in the brain, but in a mutual relationship between the

organism and environment (Jarvilehto, 1998a, 2000b; Kiverstein & Miller, 2015;

Lewis, 2005; Turvey, 2009).

14 Chapter 2: Literature review

“One doesn’t find mental activity, psyche, or emotions within the organism

(in its brain or stomach), as little as they can be found in external

stimulation. Mental activity is not activity of the brain, although the brain is

certainly an important part of the organism-environment system”


Complex dynamic systems such as humans have the capability to self-organise

their actions or behaviour to achieve specific task objectives without direct input

from higher order structures or predetermined rules (Kelso, 1995; Lewis, 2000b).

When considering a human as a complex dynamic system, the informational

variables in an environment, along with associated goal objectives and intentions

influence how an individual behaves in specific contexts. Indeed, a critical aspect of

the individual-environment relationship surrounds the perception of affordances

available to achieve a desired result or outcome (Jarvilehto, 2000a) This takes into

account the self-organising processes of affects, behaviours and cognitions in relation

to both physical and psychological activity (Kelso, 1994; Lewis, 1996). The theory

of functional systems (developed from a neurophysiological approach - see Anokhin,

1974; Egiazaryan & Konstantin, 2007) supports this integrative notion. From this

approach psychological concepts (e.g. emotion, perception, and action) are viewed as

integrated components of holistic systems that self-organise with respect to goal-

directed behaviour (Alexandrov & Jarvilehto, 1993; Anokhin, 1974; Jarvilehto,

2001). Therefore, the desired or required achievement of useful (functional)

behaviour informs the organisation of relevant and necessary system components.

States of system organisation (behaviour) that are stable and functional for a

given task are described as attractors. Attractor states are considered functional

when system components are integrated in a manner that achieves desired results


(Jarvilehto, 2001). Stable attractors are considered to have deep ‘wells’, which

represent previously learnt and commonly adopted patterns of behaviour (Kelso,

1995; Thelen & Smith, 1994). Depending on the depth or stability of an attractor,

changes in informational variables (e.g. control parameters, see Balagué, Hristovski,

Aragonés, & Tenenbaum, 2012; Passos et al., 2008) and/or task objectives have the

potential to perturb stable states of behaviour. Perturbations lead to a phase

transition in the state of the system, often producing instability. Unstable states, or

repellors, correspond to a ‘hill’ above potential ‘wells’ where behaviour is highly

variable and usually less functional. Therefore unstable states are easily influenced

by changes to informational variables both internal and external to the individual or

system.

Through the process of learning and development a previously unstable

repellor may emerge into a more stable attractor, initially as a shallow well, before

potentially progressing to a deep well (Kelso, 1995; Kelso & Tognoli, 2009;

Vallacher & Nowak, 2009). Therefore, an individual adopts a novel and functional

response, portrayed as a stable pattern of organisation that over time becomes a

characteristic response to a specific experience or environment (Lewis, 2000b;

Thelen, 2002; Thelen & Smith, 1994). Conversely, as a result of negative

experiences and/or poor performances a large repellor may develop, emerging as a

characteristic and dysfunctional response to particular situations. The role of a coach

or practitioner in sport is to recognise and manipulate key constraints in order to

perturb a dysfunctional repelllor, and subsequently drive the system (i.e. individual

learner) towards a more functional state of organisation (Davids, Araújo, Hristovski,

Passos, & Chow, 2012; Warren, 2006).


In some cases an individual may develop several coexisting attractor states that

are available to choose from, referred to as multistability. A system displaying

multistability, therefore, has the ability to switch between selected attractor states as

necessary, which can be advantageous with regards to adaptability, but also

potentially detrimental due to inconsistency in performance (Chow, et al., 2011;

Kelso, 2012; Kiverstein & Miller, 2015). A system can also display metastability,

which acts as a pivot when a system is poised to emerge into a different stable state

(i.e. multistability) (Kelso, 2012; Kelso & Tognoli, 2009). Metastability reflects the

tendency of system components competing in an attempt to function both

individually, and in unison concurrently (Kelso, 2008). Therefore metastability is a

state of partial organisation that allows a system to function while transitioning

between co-existing states of coordination (Chow, et al., 2011; Pinder, Davids, &

Renshaw, 2012).

Ecological dynamics concepts are commonly used to describe the acquisition

and coordination of physical movement behaviour, recently in relation to sport (see

Araújo, et al., 2006; Balagué, Torrents, Hristovski, Davids, & Araújo, 2013; Chow,

et al., 2011; Davids, Araújo, & Shuttleworth, 2005; Vilar, Araújo, Davids, & Button,

2012). However, many of the same principles have also been applied to describe

how individuals develop interpretations of events in relation to the self-organising

relationship between cognition and emotion (Lewis, 1996, 2000a, 2000b, 2002,

2004, 2005). While cognition and emotions are often treated as separate entities in

psychology (for an overview see Mathews & MacLeod, 1994; Pessoa, 2008; Zajonc,

1980), the three components of affect, behaviour and cognition have been shown to

influence each other to form emergent appraisals of experiences (Frijda, 1993;

Larsen & Diener, 1987; Lewis, 1996).


The process of forming a cognition-emotion appraisal can be described from

both a ‘bottom-up’ and ‘top-down’ approach (Cromwell & Panksepp, 2011; Lewis,

2004; Panksepp, 1998). The bottom-up approach involves the detection of

information relating to a task with intrinsic or learned affective value, evaluated by

structures such as the amygdala, before appropriate outputs are sent to areas

including the hypothalamus and brain stem to facilitate action tendencies (see

Pessoa, 2011). Traditionally the amygdala has been referred to in relation to fear

processing or conditioning, but more recently has also been shown to be involved

with attention, associative learning and affective value (LeDoux, 2000; Pessoa, 2008;

Zadra & Clore, 2011).

A top-down approach (often likened to the function of a computer) provides an

understanding of how humans can cognitively appraise the same situation differently

according to the situational context and their goals (Uphill, McCarthy, & Jones,

2009). From this approach informational variables are attended to through the

anticipation of emotion-laden stimuli. Therefore, humans can regulate emotion and

associated action tendencies through the use of higher cognitive processes such as

selective attention and memory. This involves taking into account the individualised

intentions and motivations towards a task (Ochsner & Gross, 2007; Tucker,

Derryberry, & Luu, 2000). The individual differences in the appraisal of the same

perceptual variables also highlight the critical role of emotion in learning or practice

environments. Every individual brings an intrinsic disposition to a task based on

previous experiences, which determine how they will interpret and approach specific

tasks.

Traditional cognitive theorists have attempted to identify specific regions of the

brain that are solely responsible for emotional responses. Alternative approaches


suggest that neural activity can be observed throughout various regions at the same

time during emotional events (Cromwell & Panksepp, 2011). For example the

limbic region (‘system’) has long been associated with emotion, but also has been

found to be involved with learning, memory, motor function, along with cognitive

and sensory processing (Pessoa, 2008). Therefore, it cannot be said for sure that

areas such as the limbic region are solely responsible for emotional responses given

the vast array of neural connections throughout the brain, and the other interrelated

functions linked to this region (Jarvilehto, 2000b; LeDoux, 2000). Of particular

importance is the finding that cognitions and emotions appear to develop in similar

areas of the brain, which highlights the close, intertwined relationship that they are

considered to share (Lewis, 2004).

The previous ideas fit well within a complex systems approach that considers

an individual as many interacting parts that self-organise under the influence of

constraints to achieve specific objectives (Kelso, 1995; Newell, 1986). From this

approach, cognition and emotion are considered to influence each other in a coupled

feedback loop (similar to perception and action) where cognitions bring about

emotions, and emotions shape cognitions (Lewis, 2004). This cyclical interaction is

considered to underpin the self-organisation of cognitive, emotional and perceptual

aspects of experiences to form Emotional Interpretations (EI) (Lewis, 1996, 2000a).

Emotional Interpretations are stable attractor states that form when emotional and

cognitive changes/responses become linked or synchronised, and subsequently

facilitate an appropriate functional action tendency (behaviour) to emerge (Lewis,

2000a). From a developmental perspective these stable attractors can be considered

as personality traits, which over time become characteristic responses to specific

experiences.


During development the relationship between cognitions and emotions in

respect to an experience may change, resulting in phase transitions from unstable to

stable states of organisation (Fogel et al., 1992; Lewis, 2004; Thelen & Smith, 1994).

Like physical behaviour, the development of EI’s may lead to metastable periods

where an individual has the potential to adopt one of a ‘cluster’ of possible

cognition-emotion states, reflecting sensitivity to changes in constraints and contexts

(Hollis, Kloos, & Van Orden, 2009; Lewis, 2000b, 2004). During a period of

metastability (i.e. learning), the behavioural tendencies would be expected to exhibit

increased variability until a more stable state of self-organisation emerges (Chow, et

al., 2011; Hollis, et al., 2009). Accompanying this variability in behaviour, variable

and individualised emotional responses also emerge reflecting the instability of a

metastable period (de Weerth & van Gert, 2000; Lewis, 1996, 2004). Much like

movement variability, emotion during learning (and performance) has been

considered, mistakenly, as unwanted noise (Davids, et al., 2003; Nesse & Ellsworth,

2009; Seifert & Davids, 2012; Smith & Thelen, 2003) Therefore sport scientists,

teachers, and coaches often attempt to remove or suppress emotion from learning and

experimental environments to facilitate controlled and predictable behaviour (e.g. de

Weerth & van Gert, 2000). Suppressing emotions during learning suggests a focus

on predictable teaching outcomes rather than allowing individuals to explore the

environment and learn from their own positive and negative experiences (Gruber,

Kogan, Quoidbach, & Mauss, 2013). By distancing emotions from learning, the role

of individual differences in learning is also diminished, preventing individuals from

finding their personal responses or solutions (Davids, et al., 2003; Seifert & Davids,

2012). Therefore removing emotional stimuli from a learning event disrupts the

crucial relationship between the individual and environment.


The role of emotions should be considered an essential part of the learning

experience because of the previously mentioned relationship between affect,

behaviour and cognition. During the appraisal of an experience the interaction

between cognition and emotion in particular is critical (Frijda, 1993). Evidence

suggests that emotions emerge subconsciously before perceptions or cognitions,

meaning that an appraisal of an experience cannot be completed without the

influence of emotion (LeDoux, 2000; Lewis, 2000a, 2005). Emotional experiences

are also considered to be vivid and memorable (Todd, Talmi, Schmitz, Susskind, &

Anderson, 2012), influential on attentional processes (Padmala & Pessoa, 2008;

Pessoa, 2011), and associated with increased arousal (Anderson, Yamaguchi,

Grabski, & Lacka, 2006). Therefore, emotions should always be considered as a

primary influence or precursor of learning experiences.

A working definition of emotion

Evidently what constitutes an emotion is difficult to conclusively define and

measure. Appendix A provides a selection of emotion definitions representing

several interpretations spanning across hundreds of years (see page 257).

Incorporating key aspects of these interpretations, the following working definition

of emotion is presented reflecting the approach adopted throughout this thesis.

Drawing prominently on ecological dynamics concepts, this working definition

outlines the overarching perspective of emotion discussed in subsequent sections and

chapters.

Emotions are dynamic self-organising states (intertwined with actions and

cognitions) that can constrain, describe, and predict emergent behaviour

across interacting time scales, with respect to the reciprocal individual-

environment relationship.


Examining the definition more closely reveals several key concepts that have

been discussed previously. The self-organising nature of emotions, actions, and

cognitions is integral to the definition in that these domains must be considered in

unison rather than as separate entities. Next, the ability of emotions to constrain,

describe, and predict emergent behaviour is recognised, linking with the two

principles of Affective Learning Design introduced in Chapter 3. In short these

principles recognise the role of emotion-laden environments in shaping emergent

behaviour, and also the potential to interpret emergent behaviour through the

consideration of emotional responses (along with actions and cognitions). The

importance of considering emotions over various interacting time scales reflects the

dynamic moment-to-moment nature of emotion. Recognising the reciprocal

relationship between individuals and the environment forms the final, and most

crucial, aspect of the working definition. This conceptualisation is a key tenet of this

principled approach, where emotions emerge through the mutual individual-

environment duality, rather than being considered exclusively ‘internal’ to an

individual (see, Bruineberg & Rietveld, 2014; Davids & Araújo, 2010; Jarvilehto,

1998a; Kiverstein & Miller, 2015; Lewis, 2004).

Emotions and human behaviour

This section discusses the scope of previous literature studying emotion, with

an emphasis on sport. In particular, the popularised and frequently studied emotional

constructs of anxiety and fear are reviewed. This overemphasis on emotions with

negative connotations is then highlighted and contrasted with work incorporating

both positive and negatively valenced emotions.


Anxiety

Anxiety is arguably the most frequently investigated and discussed emotion,

particularly in the context of sport (Hanin, 2000b). State anxiety refers to the

moment – to – moment, short time scale fluctuations in feelings of apprehension and

tension with associated arousal of the autonomic nervous system (Martens, Vealey,

& Burton, 1990; Spielberger, 1966). Trait anxiety is a more permanent part of an

individual’s personality that has developed over time and reflects acquired

behavioural tendencies (Spielberger, 1966; Weinberg & Gould, 2015). Anxiety is

further categorised as either cognitive or somatic reflecting the multi-dimensional

characteristics of the broad construct (Borkovek, 1976; Davidson & Schwartz, 1976).

Cognitive anxiety represents the worries, negative thoughts, and concerns about

performance. Somatic anxiety reflects perceptions of physiological (arousal) and

affective changes produced by the autonomic system (Gould & Krane, 1992;

Martens, Vealey, et al., 1990). To conceptualise the relationship between anxiety

and sport performance a range of models and theories have been proposed based on

increasing levels of physiological arousal, for example Drive Theory, and the

Inverted-U hypothesis (for overviews see - Gould & Krane, 1992; Martens, Vealey,

et al., 1990; Weinberg & Gould, 2015). A related model that incorporates ideas

presented throughout this thesis is the Cusp Catastrophe Model (Fazey & Hardy,

1988; Hardy & Fazey, 1987). This model recognises the multi-dimensional factors

of physiological arousal (somatic anxiety) and cognitive anxiety on performance,

whereby these variables are facilitative of successful performance up to a specific

threshold, but detrimental beyond the threshold. Therefore, performance is deemed

to be dependent on the intensity of these linked control parameters (arousal,


cognitive anxiety), which have the potential to destabilise system behaviour when

increased beyond individualised thresholds (Hardy, 1996; Newell, et al., 2001).

The influence of anxiety on psychological processes in a range of interceptive

actions has been widely investigated by a number of researchers. For example, the

influence of induced anxiety on attentional processes in driving tasks has been

examined to determine how performance and peripheral awareness might be affected

(Janelle, Singer, & Williams, 1999). Participants were required to complete a

primary task of driving a race car simulator while also attending to peripheral stimuli

and distractions in the form of flashing lights. Performance anxiety was created by

designating sessions that were for familiarisation (low anxiety), practice (moderate

anxiety) or competition (high anxiety). Results revealed that in conditions of higher

anxiety participants were less accurate and slower to respond to the peripheral lights,

in addition to demonstrating decreased performance in the primary driving task. In

conditions of high anxiety and peripheral distraction the visual search tendencies of

participants were found to be erratic with more fixations to peripheral locations.

This was deemed to be as a result of attentional narrowing (‘tunnel vision’), where

heightened anxiety and/or arousal restricts the attention to relevant stimuli in the

environment, particularly in dual tasking activities (Easterbrook, 1959). Therefore

the emotional responses, in the form of increased anxiety in relation to a pending task

(see Öhman, 2008), influenced the cognitive attentional processes and subsequently

impacted performance. Similarly, anxiety was found to influence the perception of

affordances for simple reaching, grasping and passing tasks (Graydon, Linkenauger,

Teachman, & Proffitt, 2012). Participants in anxiety conditions were found to

display more conservative behaviour traits (avoidance, freezing) which were

detrimental to their perceived performance capabilities.


Anxiety has also been found to influence the performance of individuals in

complex movement tasks in sport. Anxiety was manipulated in wall climbing

through the use of identical climbing routes (traverses) of different heights (Pijpers,

Oudejans, Bakker, & Beek, 2006). In three related experiments the higher route (i.e.

high anxiety) was found to produce differences in perception and performance when

compared with the lower route (low anxiety). Both perceived and actual maximal

overhead reaching height were found to be reduced in the high anxiety condition

indicating that the higher traverse influenced the perceived action capabilities of the

climbers. Secondly, an increase in the number of holds used in the high anxiety

condition was reported, suggesting that the increased emotion intensity (anxiety)

influenced the climbers in terms of being more conservative in the manner they

completed the course. The third experiment required climbers to detect lights

projected around the climbing surface. Results revealed that in the high anxiety

condition there were fewer lights detected and therefore it can be assumed that the

climbers narrowed their attention as a result of anxiety. Closely related studies also

found that high anxiety was responsible for physiological responses including

increases in heart rate, muscle fatigue, and blood lactate. Movement behaviour was

also found to be influenced by higher anxiety conditions through measures such as

hold time, course completion time, and exploratory movements (Pijpers, Oudejans,

& Bakker, 2005; Pijpers, Oudejans, Holsheimer, & Bakker, 2003). Hence, the

manipulation of anxiety through different traverse heights was found to influence the

intensity of performance and the affordances perceived, and acted on, by the

climbers. Height was also used as a method for increasing anxiety when

investigating the influence of anxiety, dual tasking, and expertise on dart throwing

performance (Nibbeling, Oudejans, & Daanen, 2012). Here the performance of


novices in particular was found to decrease in high anxiety conditions. High anxiety

also led to reduced performance in the secondary cognitive task in the form of less

frequent responses to a counting task. Finally, gaze fixations on the target were

found to be shorter in length and moved away from the target earlier in the high

anxiety conditions.

Fear

Closely related to anxiety is the emotional state of fear, which motivates the

need to escape or avoid situations and objects that are perceived as dangerous or

threatening (Frijda, 1986; Öhman, 2008; Öhman & Mineka, 2001). Throughout

mammalian evolution, responses to fearful stimuli (e.g. predators) have included

escaping, freezing, and attacking depending on the constraints of the environment

(Blanchard & Blanchard, 1988). Fear is also an influential factor on human

performance in sporting contexts where there is the potential for physical injury. For

example, the ability of gymnasts to overcome the fear of suffering injury (and re-

injury) has been examined to determine psychological and emotional factors that

influence their continued performance (Chase, Magyar, & Drake, 2005; Mace,

Eastman, & Carroll, 1986). Furthermore, by manipulating the height of balance

beams the interaction between functional and dysfunctional emotions in performance

was investigated (Cottyn, De Clercq, Crombez, & Lenoir, 2012). For the higher

beam heights, dysfunctional emotions (i.e., fear, anxiety) were more evident and

resulted in decreased performance, particularly on the first attempt. These findings

suggest that manipulating the perceived difficulty and/or intensity of a performance

task has the potential to influence the emotional tendencies of performers. In the

next section, the discussion is broadened to consider the influence of positive and

negative emotions on performance.


Positive and negative emotions

The emotions of fear and anxiety are generally viewed as negative and are the

most commonly studied emotions in relation to human performance because of the

evolutionary significance of identifying objects or experiencing events that are

potentially threatening (Burgdorf & Panksepp, 2006; Öhman, Flykt, & Esteves,

2001; Woodman et al., 2009). The influence of positive emotions (e.g. happiness,

joy) has also been investigated, often in direct comparison with negative emotions.

For example, when positive, negative, and neutral photographs were shown to

participants, the photographs eliciting positive emotions were found to be most

memorable in a recall task (Becker, 2012). Further investigation of emotions other

than fear and anxiety suggest that positive emotions can also influence performance

(Raedeke, 2007; Woodman, et al., 2009). Positive emotions have been linked with

widening the scope of attention, and behavioural tendencies (thought-action

repertoires) in a video based investigation including positive, neutral, and negative

visual stimuli (Fredrickson & Branigan, 2005; Fredrickson & Losada, 2005). This is

in congruence with early ideas of James (1884) who outlined that emotions should be

considered as valenced, relating to either positive or negative experiences (Oatley &

Jenkins, 1992). On the other hand it is difficult to categorically specify that emotions

can be described as positive/negative, or functional/dysfunctional because of

individual differences in how emotion-laden environments are perceived (Bhalla &

Proffitt, 1999; Friesen et al., 2013; Nesse & Ellsworth, 2009; Proffitt, Bhalla,

Gossweiler, & Midgett, 1995). For example, anxiety might be perceived as negative

by one individual and therefore dysfunctional, a second individual might find

increased anxiety to be functional for performance and therefore a positive (Fazey &

Hardy, 1988; Hardy, 1996). Some have considered positive and negative items to be


opposing extremes of the same emotion continuum (Diener, Larsen, Levine, &

Emmons, 1985). Investigations comparing the frequency and intensity of contrasting

emotions suggest otherwise, advocating that positive and negative emotions should

be treated independently (Carver & Scheier, 1990; Diener, et al., 1985; Warr, Barter,

& Brownbridge, 1983).

From the perspective of the organism-environment theory put forward by

Jarvilehto (1998a), positive and negative emotions are referred to in terms of

different states of system organisation relating to the achievement of results

(Jarvilehto, 1998a, 2000b; Turvey, 2009). Positive emotions indicate transitions in

the organism-environment system organisation involving the integration of new

experiences relating to successful action or results (Jarvilehto, 2000a; Spinoza,

1674/2006). The combination of successful results and positive emotions

accompanies the development of a system to new states of organisation through

differentiation (i.e. learning) (Jarvilehto, 1999, 2000b, 2009). On the other hand,

negative emotions represent situations where a desired result is not achieved and the

system disintegrates (reorganises) to a prior functional state (Jarvilehto, 2000a,

2000b; Spinoza, 1674/2006). This eventuality indicates that a key component of the

organism-environment system is ‘missing’, which acts as a rate limiter (see Thelen,

1995) or barrier for the transition and development to new functional states. Self-

organisation (integration – disintegration) of a system constantly takes place with the

ever present interaction between emotion and action, or as stated by Jarvilehto

(2000b, p. 43) “there is no action without emotions”. Therefore it is possible that an

organism-environment system might be in a state of partial (i.e. metastable)

integration/disintegration exhibiting positive and negative emotions in unison

(Jarvilehto, 2000a).


Intentionality

The concept of intentionality refers to the ‘aboutness’ of and reason for some

process, act or experience in relation to an object or event (Searle, 1983; Shaw,

2001). All of the concepts and examples discussed below are in some way related to

intentionality, because they are directed at or about a property of an environment,

whether it is a previous performance, or an object that is feared. Intentionality can be

considered to focus or concentrate the attention of an individual on situations that are

perceived to be potentially beneficial or harmful (Bruineberg & Rietveld, 2014;

Lewis, 2004). Emotion and intentionality must therefore be considered alongside

each other on account of emotions inherently being ‘about’ or ‘toward’ something

(Dewey, 1895; Jarvilehto, 2001). For example, an emotion might be about or

towards a specific object, event, situation or performance outcome that is deemed to

be of some value to the individual across interacting time scales (Bruineberg &

Rietveld, 2014; Lewis, 2002; Newell, et al., 2001). Jarvilehto (2000a, p. 56)

suggested that “emotion could be very generally defined as a process of

reorganisation (integration – disintegration) of the organism-environment system,

expressed most clearly in relation to the realisation of the expected behavioural

result”. In other words, emergent emotions about an event or situation reflect an

individual’s evaluation of how their current capabilities meet their goals or

expectations (Jarvilehto, 2001).

Considering the influence of intentionality on practice and learning tasks

should not be underestimated. Designing tasks that involve objectives, goals and

aims provide some context to a learning task and subsequently a reason or motive for

participation (Araújo, et al., 2006; Shaw & Turvey, 1999). For example,

manipulating instructional constraints in basketball (i.e. time and score) to


manipulate scoring pressure introduced an intentionality for both attacking and

defending players, representative of situations experienced in a game (Cordovil et

al., 2009). Similarly, through the manipulation of the proximity to goal in

association football, the intentions (strategies, tactics) of players were found to differ

reflecting the situation specific information that was included (Headrick et al., 2012).

As a result, the performers in the two given examples have been given some context

for action, either through explicit instructions or by task constraints defining their

purpose in the specific situation. As a result, an ‘engagement’ with the performance

environment is maintained reflecting the key link between perception and action in

complex systems (Kiverstein & Miller, 2015; Lewis, 2005; Solomon, 2008). These

ideas have clear implications for the design of experiments and practical activities in

sport and exercise.

Representative design

Based on the relationship between perception (information) and action

(movement) it is imperative to consider Egon Brunswik’s (1956) concept of

representative design when designing tasks for experimental, learning and practice

purposes (Araújo, et al., 2007). Brunswik advocated the study of organism-

environment relations in psychology through the preservation of stimuli that would

naturally be available to a performer/participant in an environment (Dhami, et al.,

2004). Furthermore, Brunswik (1955) proposed that the tendency of (at the time

popular) systematic design to isolate and manipulate individual variables or stimuli

destroyed the organic relationship between perceptual information offered by the

environment and the actions of a performer. Systematic design focussed on

maintaining a high internal validity by identifying causal relationships between

isolated variables. However, this was often to the detriment of external validity,


where the finding may not be generalised when other confounding variables are

reintroduced (Dhami, et al., 2004; Pinder, Davids, et al., 2011b). A term often

confused with external validity and representative design is another Brunswikian

concept – ecological validity (see Araújo, et al., 2007; Hammond, 1998; Pinder,

Davids, et al., 2011b). Ecological validity refers to the statistical correlation between

perceptual cues (proximal effects) present in the environment and an eventual (distal)

outcome state. In other words, the likelihood of proximal variables inferring a distal

outcome, such as deciding whether a criminal may reoffend (distal) based on their

previous actions (proximal) (Dhami, et al., 2004). Consequently, ecological validity

refers to the reliability of a perceptual cue, compared with representative design

where findings pertaining to cues can be generalised in settings outside of a

laboratory, such as sport (Kirlik, 2009).

Representative learning design

To increase the relevance of Brunswik’s (1956) concept of representative

design for use by sport scientists, coaches and physical educators, Pinder et al.

(2011b) introduced the term representative learning design. The overall premise of

representative learning design (RLD) is to consider the inclusion of representative

constraints in sport performance simulations, and thereby evaluate the degree to

which practice and learning tasks simulate a particular environment (Araújo, et al.,

2007; Pinder, Renshaw, Davids, & Kerhervé, 2011). Key aspects deemed to uphold

representative learning design are the functionality and action fidelity of practice or

learning contexts. Functionality refers to the performer being able to complete

specific behaviours (i.e. decision making, movement coordination) in both practice

and performance environments because of direct comparisons between information

available in each context. Action fidelity refers to the level of transfer in


performance from a simulation (e.g. learning or practice tasks in sport) to the

simulated system (e.g. sport performance/competition), originating from the study of

flight simulators (Stoffregen, Bardy, Smart, & Pagulayan, 2003). Therefore, practice

and learning tasks which are high in action fidelity must provide conditions that

accurately simulate the performance environment of interest (Pinder et al., 2012). As

an example, the use of dynamic game-based training in team ball sports provides

practice conditions akin to competition, rather than highly constrained and structured

‘drills’ that decompose the perception-action relationships experienced in match

contexts (Chow, et al., 2007; Renshaw, Chow, Davids, & Hammond, 2010).

Currently, the literature discussing representative (learning) design in sport has

focused on physical and visual informational variables of a practice environment and

how they relate to actual match or competition contexts (Pinder, et al., 2014). One

example is the comparison of live versus simulated visual information in cricket

batting to establish the representativeness of popular practice methods (Pinder,

Davids, et al., 2011a). Here the movement organisation of cricket batters was

compared when batting against real ‘live’ bowlers, video simulations of the same

bowlers, and a ball projection machine. Results revealed that the movement

characteristics of batters when facing the live bowlers were more comparable with

the video simulations than with the ball projection machine, positing that the

availability of visual information from the bowlers’ action in the video condition

provided a more representative simulation. Video simulation and live (in situ)

conditions were also used to examine football goalkeeper gaze and movement

behaviour under a range of conditions (Dicks, Button, & Davids, 2010).

Goalkeepers were found to spend more time fixating on the body movements of

penalty takers rather than the ball in conditions where responses were restricted to


little or no movement (i.e. verbal, joystick and simplified body movements). When

asked to physically intercept penalty kicks, the goalkeepers fixated on the ball and

penalty taker for similar periods of time. These findings illustrate that different gaze

behaviour was evident when the goalkeeper was required to act as they would in a

match compared with when merely perceiving information for restricted movements

or verbal responses that are not representative of responses required in a match.

Through different task vehicles and scales of analysis the previous examples

summarise some recent findings and approaches regarding the representativeness of

sport practice/performance settings. These studies provide clear implications for

upholding the representativeness of physical and visual variables in practice

environments, therefore allowing performers to perceive affordances from

informational variables that they would expect to encounter in match or competition

settings. However, while sampling environmental information is a good start,

through the following sections it will be argued that the influence of emotion on

learning and performance should also be considered when designing learning

environments.

Learning and performance

Motor learning and performance are difficult to separate from each other, with

learning processes derived from observable performances typically under different

and highly controlled conditions, such as learning to perform (Schmidt & Wrisberg,

2008). From a complex systems approach, learning can be described as the transition

to a preferred functional movement pattern (attractor) from a previously learnt

movement pattern (Chow, Davids, Button, & Rein, 2008; Liu, Mayer-Kress, &

Newell, 2006; Thelen, 1995). Learning and development are often discussed


alongside each other as distinct, yet integrated concepts (Newell, et al., 2001). In

general, learning relates to changes (e.g. performance, ability) over short time scales

while development describes changes over longer periods of time (Granott &

Parziale, 2002; Newell, et al., 2001). As learning is inferred from performance,

learning and experimental environments should sample the conditions of the relevant

performance environment to uphold representative learning design (Pinder, Davids,

et al., 2011b).

Learning takes place over different interacting time scales that shape how an

individual might approach a task (Newell, et al., 2001). The different time scales are

often described as real and developmental, or micro and macrodevelopment,

respectively (Granott & Parziale, 2002; Lewis, 2002; Thelen, 1995). From a

complex systems perspective real time events shape long term developmental

patterns or states, and conversely the established developmental states influence

learning and microdevelopment in real time (Lewis, 2000a, 2002). These reciprocal

influences between micro and macrodevelopment have been described as ‘cascading

constraints’ on learning and performance (Lewis, 1997, 2002). Therefore,

established emotional interpretations (EI) of an experience influence how an

individual will approach learning a new task and in turn the process of learning will

modify the overall EI (Lewis, 2000a, 2004). Taking a practical perspective to this

concept highlights the critical role of an initial experience (e.g. first teacher or coach)

for shaping an individual’s approach to learning, as an early EI constrains subsequent

learning experiences. For example, if a performer has few opportunities to be

involved in initial training sessions, they are not likely to become engaged in future

activities based on these prior experiences.


Goal orientation

The emergence of positive (e.g. joy, love, happiness) and negative (e.g. fear,

anxiety, anger) emotions must be considered in relation to the goals an individual has

for specific tasks. In a study on achievement tasks in children with equal abilities,

different emotional responses were found to coincide with the outcome of failure,

namely mastery and helpless responses (Diener & Dweck, 1978, 1980). Children

displaying an adaptive mastery response displayed persistence, pursued the challenge

and related failure to insufficient effort, while those exhibiting a helpless response

were observed to be maladaptive, avoided the challenge and related failure to

insufficient ability (Dweck, 1986; Elliot, 2005; Elliott & Dweck, 1988). To

determine why individuals of equal ability performed in such different ways the

concept of goals was introduced, where the goals of an individual create the

framework for their performance (Dweck & Elliott, 1983; Dweck & Leggett, 1988).

Various achievement goal theorists have commonly defined two distinct types of

goals, for example, performance and learning goals (Dweck, 1986), ego and task

involvement goals (Nicholls, 1984), amongst others. Despite subtly different labels

and definitions, two types of goals can be distinguished. Performance goals relate to

being successful, and being judged by others as competent, while mastery goals refer

to the development of skills and competency (Ames & Archer, 1987; Elliot, 2005).

Achievement motivation theorists have for many years incorporated an approach-

avoidance distinction in the study of competence in human behaviour (Atkinson,

1957; McClelland, Atkinson, Clark, & Lowell, 1953). Competence can be defined as

one’s mastery of a task in relation to the task requirements, an individual’s own

previous and potential standard, and in comparison to the standard of others (Elliot,

2005). The approach-avoidance concept describes the valence of competence


whereby individuals are motivated to either approach success and demonstrate

competence, or attempt to avoid failure and the display of incompetence (Oatley &

Jenkins, 1992).

Several studies have investigated the role of goals and an individual’s approach

to emotion-laden tasks. For example, the perspectives of extreme sports (e.g. BASE

jumping, big wave surfing) participants were studied to understand how individuals

perceive such tasks and overcome the fear of failure, which would almost certainly

result in serious injury or death (Brymer & Oades, 2009). In this case, fear was

found to be interpreted and experienced in a way that was meaningful and

constructive for performance and continued participation (Brymer & Schweitzer,

2013). In a less extreme example, youth athletes from various sports were

questioned to determine why younger athletes may fear failure in the achievement

orientated setting of various sports (Sagar, Lavallee, & Spray, 2007). The most

frequent consequences of failure feared by athletes were diminished perception of

self (e.g. decreased confidence), no sense of achievement (e.g. losing), and the

emotional cost of failure (e.g. guilt, criticism from others). Finally, the goal

involvement states of Judokas (Judo competitors) were observed during a practice

bout to determine whether the goal orientations fluctuated during performance

(Gernigon, d'Arripe-Longueville, Delignieres, & Ninot, 2004). Through both

quantitative and qualitative analysis techniques the Judokas were found to frequently

alter their goal orientations depending on the various properties of the dynamic

system (e.g. opponent’s actions, success of attacks). This study was particularly

interesting in that it measured goal states throughout the performance; an uncommon

methodology in many of the studies of psychological factors in sport performance.

Goals have also been found to differ between competition and training contexts in


golf putting and football (soccer), with the overall finding of higher task (mastery)

orientation during training and increased ego (performance) orientation in

competition (van de Pol, Kavussanu, & Ring, 2012a, 2012b). This is a critical

finding in relation to the design of practice environments as it details that individuals

approach the two contexts very differently. Therefore, the distinction between goals

associated with different learning and performance tasks is a critical consideration

for this project.


Motivation

As action tendencies, motivational consequences may lead a performer toward

(motivated) or away (unmotivated) from an object, or task (Jones, 2003). Self-

determination theory (Deci & Ryan, 1985, 1991) states that performers engage in

tasks for reasons, which originate from internal or external influences. Internal

motivations are more self-deterministic in comparison with external motivations that

are influenced by outside sources (e.g. coach or parent), however both varieties

interact to develop individualised characteristics and interpretations of experiences

(Vallacher & Nowak, 2009). Hence from this perspective, the two basic types of

motivation are intrinsic and extrinsic motivation (Deci & Ryan, 1985). Intrinsic

motivation refers to participating in an activity for the inherent satisfaction, fun or

challenge involved without the influence of external pressure, rewards, instructions

and consequences. In other words intrinsic motivation represents the ‘free choice’ of

an individual to be involved in an activity or task (Deci, 1971; Ryan & Deci, 2000a).

Conversely, extrinsic motivation occurs when an activity is performed to

achieve an outcome other than for pure enjoyment or satisfaction. Consequently, the

majority of day-to-day activities such as going to work are extrinsically motivated

because they pertain to a separate outcome (i.e. earning a living). The completion of

some tasks may be both intrinsically and extrinsically motivated, for example going

to work may be a source of enjoyment (intrinsic), while also a means to earn an

income (extrinsic). Ryan and Deci (1985; 2000a) defined the key regulators of

extrinsic motivation as external regulation, introjection, identification, and

integration (often not discussed, see Vallerand & Blanchard, 2000). External

regulation concerns the rewards or punishment associated with completing a task (or

not). The ego involvement (see Nicholls, 1984) of a performer is described by


introjected regulation whereby a task is completed to uphold or improve self-esteem,

whilst also avoiding anxiety or guilt. Identification refers to a performer consciously

recognising and accepting the value of an activity in relation to his or her goals.

Integrated regulation occurs when tasks and goals are fully incorporated with

existing values and needs of a performer much like intrinsic motivation, however

tasks are still completed for the value of their outcome rather than pure enjoyment.

Finally, the term amotivation describes the motivational state of a performer when

they lack any intention to act from either intrinsic or extrinsic influences. Therefore

a low-to-high continuum of self-determination begins with amotivation increasing

through external regulation, introjected regulation, identified regulation, integrated

regulation, and intrinsic motivation (Deci & Ryan, 1985, 1991; Ryan & Deci,

2000a). By considering emotion as a key aspect of representative learning design,

the self-determined motivation of a performer to engage in a practice environment

may be improved.


Summary

The previous sections have discussed the influence of emotion on human

behaviour and advocated for the recognition of emotion as an integral feature of

learning events. Emanating from dynamic systems and ecological psychology

concepts a principled ecological dynamics approach to emotion during learning, has

been put forward, representing the theoretical background of this PhD. This

approach recognises the key relationship between an individual and the environment

for the detection of information that is perceived in order to afford goal orientated

behaviour. Several substantial gaps in the current literature have been identified,

highlighting that a principled approach to emotions during learning in sport is

severely lacking. With this is mind it was proposed that the concept and practice of

representative learning design could be enhanced by embracing emotions as part of

learning events. To provide background and rationale for investigating these issues,

evidence was presented detailing the role of emotion in learning and performance,

and reasons why it may have received little attention. This included discussion of the

relationship between affect, behaviour (perception-action), and cognition from a

systems perspective to reveal how these three domains influence each other.

Drawing from the interaction between affect, behaviour, and cognition, in-depth

individualised approaches were advocated, incorporating the influence of interaction

time scales of learning. Following this, a range of examples highlighted how

emotion and interrelated psychological concepts (e.g. attention, goal orientations, and

motivation) influence behavioural tendencies of individuals. When combined with a

principled theoretical approach, these practical findings highlight potential research

directions based on creating and studying emotion-laden learning environments in

sport.

Chapter 3: Affective learning design: Developing a principled approach to emotion in learning 41

Chapter 3: Affective learning design: Developing a principled

approach to emotion in learning

Through the review of literature in Chapter 2 it was identified that a principled

approach to the role of emotion in learning during sport was lacking. This chapter

incorporates and builds on the reviewed literature to develop principles of Affective

Learning Design, underpinned by Representative Learning Design (Pinder, Davids,

et al., 2011b) and Ecological Dynamics concepts. Affective learning design (ALD)

will be referred to throughout the following chapters as it represents the key

theoretical premise of the PhD programme.

This chapter is based on the following peer reviewed article:


dynamics of expertise acquisition in sport: The role of affective learning design.

Psychology of Sport and Exercise, 16, 83-90. doi: 10.1016/j.psychsport.2014.08.006

42 Chapter 3: Affective learning design: Developing a principled approach to emotion in learning

Abstract

This chapter discusses the role of affect in designing learning experiences to

enhance expertise acquisition in sport. The design of learning environments and

athlete development programmes are predicated on the successful sampling and

simulation of competitive performance conditions during practice. This premise is

captured by the concept of representative learning design, founded on an ecological

dynamics approach to developing skill in sport, and based on the individual-

environment relationship. Here it is discussed how the effective development of

expertise in sport could be enhanced by the consideration of affective constraints in

the representative design of learning experiences.

Based on previous theoretical modelling and practical examples two key

principles of Affective Learning Design are proposed: (i) the design of emotion-laden

learning experiences that effectively simulate the constraints of performance

environments in sport; (ii) recognising individualised emotional and coordination

tendencies that are associated with different periods of learning. Considering the

role of affect in learning environments has clear implications for how sport

psychologists, athletes and coaches might collaborate to enhance the acquisition of

expertise in sport.


Introduction

In sport, performers must adapt to the constraints of dynamic performance

environments, with commensurate variable conditions and situations, while

performing under different emotional states that constrain their cognitions,

perceptions and actions (Jones, 2003; Lewis, 2004). Despite the documented

presence of emotions1 in sport performance, thus far (see Vallerand & Blanchard,

2000), limited attention has been paid to the role that emotions might play during the

acquisition and development of expertise. Traditionally, emotions have generally

been viewed as negative and detrimental constraints on behaviour, considered better

to be removed from practice task contexts until a skill is well established (Hutto,

2012). This reductionist approach to learning design is in line with traditional

thinking in the acquisition of skill in which practice tasks are decomposed to

putatively reduce the cognitive loading on performers as they attempt to enhance

expertise (Lewis & Granic, 2000). Here questions are raised on the reductionist

approach to learning design and discuss an alternate principled approach suggesting

how affective constraints on behaviour may be included during the acquisition of

expertise in sport, drawing on the theoretical rationale of ecological dynamics.

In existing research on movement behaviours, ideas from dynamical systems

theory have been integrated with concepts from Gibsonian ecological psychology,

forming the ecological dynamics approach to understanding performance and

learning (Araújo, et al., 2006; Davids, Williams, Button, & Court, 2001; Warren,

1 The broad term of affect refers to a range of phenomena such as feelings, emotions, moods, and

personality traits that interact over different time scales. Here affect will be used interchangeably with

emotion to follow previous modelling of cognition, emotion and action (Lewis, 2000a; Vallerand &

Blanchard, 2000).


2006). An ecological dynamics approach to enhancing expertise recognises the need

for individuals to form mutual functional relationships with specific performance

environments during practice and training (Araújo & Davids, 2011; Davids, Araújo,

Vilar, Renshaw, & Pinder, 2013; Seifert, Button, et al., 2013). In a functionalist

approach to the study of perception and action, Gibson (1979) emphasised the role of

the environment and proposed that an individual's movements bring about changes in

informational variables from which affordances (invitations for action) are perceived

to support behaviours (Withagen, de Poel, Araújo, & Pepping, 2012). As a result a

cyclic process is created where action and perception underpin goal-directed

behaviours in specific performance environments (Gibson, 1979).

Studying emergent behaviours and the acquisition of expertise at this

individual-environment scale of analysis takes into account how perceptions, actions,

intentions, feelings and thoughts continuously emerge under the constraints of

information external and internal to the individual (Seifert & Davids, 2012; Warren,

2006). Humans, conceptualised as complex dynamic systems, exhibit self-

organising coordination tendencies during learning and performance to achieve

specific task objectives (Kelso, 1995; Lewis, 2000b). The informational variables in

a specific performance environment, along with associated goals and intentions,

constrain how each individual behaves (Davids, et al., 2001; Freeman, 2000;

Juarrero, 2000). Coordination tendencies (e.g., behaviours in human movement

systems) that are stable are described as attractors (Kelso, 1995; Zanone & Kelso,

1992). Stable attractors are states of system organisation that represent well learned,

stable patterns of behaviour (Kelso, 1995; Thelen & Smith, 1994). It is important to

note that coordination tendencies may be functional or dysfunctional in terms of

meeting the demands of a specific task, or during learning (Warren, 2006).


Depending on the depth or stability of an attractor, changes in informational

variables that act as control parameters have the potential to perturb (disrupt)

coordination tendencies (e.g., Balagué, et al., 2012; Passos, et al., 2008).

Perturbations can lead to phase transitions in coordination tendencies, often

producing changes in behaviour. Unstable system states correspond to a ‘hill’ above

potential ‘wells’ where coordination tendencies may be variable and possibly less

functional (Kelso, 1995; Vallacher & Nowak, 2009). Unstable system states are

more open to influence by changes to informational variables both internal and

external to an individual during performance (Davids, et al., 2001). Through practice

and experience in sport, athletes, considered as dynamic movement systems, can

learn to enhance stability of performance behaviours and increase their resistance to

perturbations, including negative thoughts, and emotions (e.g., differences in ice

climbing performance between experts and novices, see Seifert, Button, et al., 2013;

Seifert et al., 2013). An important question for sport psychologists and coaches

concerns how practice programmes can be designed to provide athletes with learning

experiences that help them to exploit functional coordination tendencies (i.e. system

states which are stable yet adaptable) under the affective constraints of sport

performance.

Ecological dynamics is an integrated theoretical rationale of human behaviour

that can underpin a principled approach to learning design in clinical (Newell &

Valvano, 1998) and sport performance environments (Araújo, et al., 2006). The basis

of behavioural change through learning involves the systematic identification and

manipulation of system control parameters (informational constraints) to perturb

stable states of organisation and facilitate transitions to more functional system states

(Kelso, 1995, 2012). Attractors can take the form of intentions, and/or goals that a


performer is 'attracted to' following changes in values of system control parameters

(Davids, et al., 2013; Davids, et al., 2001; Warren, 2006). Stable system states often

represent desired forms of organisation that are functional. Enhanced functionality,

i.e. ‘what works’ (see Thelen & Smith, 1994), is achieved when an athlete establishes

a successful relationship with a performance environment and individualised task

goals are achieved (e.g., through more accurate or faster performance outcomes).

Simultaneously, functional coordination tendencies can satisfy the psychological

needs (i.e. ‘what feels good’) of each individual performer in particular performance

situations (Carver, Sutton, & Scheier, 2000; Hollis, et al., 2009; Lewis, 2004). In

order for a behavioural attractor to become stable through learning, the intrinsic

dynamics (the predispositions and tendencies) of each performer and the task

dynamics (e.g., specific performance requirements) must converge (Davids, et al.,

2001; Zanone & Kelso, 1992). The relative stability of behavioural attractors is

important to facilitate achievement of successful performance at specific points in

time. But, learning environments also need to be dynamic and variable to allow an

individual to adapt to changing individual, task and environmental constraints over

the short and long time scales of development (Lewis, 2002; Newell, 1986). A key

task for sport psychologists and practitioners is to understand how to effectively

manipulate constraints to facilitate the development of new behavioural attractor

patterns essential for expertise acquisition.

Sport psychologists have begun to identify control parameters to design

effective learning environments that are carefully matched to each individual’s

intrinsic dynamics, or predispositional behavioural tendencies. Carefully designed

learning environments can guide athletes towards metastable performance regions, in

which a functional blend of coordination stability and adaptability can result in rich


behavioural solutions emerging (Hristovski, Davids, Araújo, & Button, 2006; Pinder,

et al., 2012). Metastability is a state of partial organisation where a system ‘hovers’

in a state of dynamic stability, switching between functional states of organisation in

response to changing constraints, and displaying subsequent behavioural flexibility

(variability, instability) (Fingelkurts & Fingelkurts, 2004; Phillips, Davids, Araújo, &

Renshaw, 2014). Metastability allows a system to transit rapidly between co-existing

functional states of organisation, essential for adaptive performance behaviours in

dynamic environments (Chow, et al., 2011; Kelso, 2012; Kelso & Tognoli, 2009).

During learning events in specific performance environments, being in a state of

metastability allows performers to discover and explore performance solutions

(Kelso, 1995; Seifert, Button, et al., 2013). In sport, empirical data has revealed how

locating samples of boxers and cricketers in metastable performance regions during

practice helped them to explore and exploit rich and creative performance solutions

to achieve their task goals (Hristovski, et al., 2006; Pinder, et al., 2012).

Adopting novel and potentially functional states of system organisation is a

consequence of learning and/or development, as individuals transit from the ‘known’

to the ‘unknown’, i.e., moving from a familiar task or situation to one that is new or

different. Of interest to sport psychologists is that increases in movement variability

during phases of learning are often accompanied by increased intensity and range of

emotions (Lewis, 2004, 2005). These emotions can be attributed to: (i) the challenges

of learning a new movement pattern; (ii) the perceived risk of failure to achieve

specific performance outcomes; and (iii), the underlying uncertainty and/or

excitement associated with performing in an unknown situation. For example, the

changes in infant sensorimotor and emotion regulation during developmental

transition periods (Lewis & Cook, 2007). Observable changes in behaviours and


emotions of athletes are of importance since they can act as predictors for potential

phase transitions in system behaviours, such as coordinated movement response

characteristics (Chow, et al., 2011; Kelso, 1995). The theoretical rationale of

ecological dynamics suggests that it is essential to design learning environments that

guide athletes towards metastable regions of a perceptual-motor workspace during

performance (physically and emotionally) to aid the acquisition of expertise in sport

(Oudejans & Pijpers, 2009; Pinder, et al., 2012). In achieving this aim, an important

challenge for sport psychologists and practitioners is how to design learning

environments that successfully simulate key constraints of competitive performance

environments in sport. Egon Brunswik (1956) advocated that, for the study of

individual-environment relations, cues or perceptual variables should be sampled

from an organism’s environment to be representative of the environmental stimuli

that they are adapted from, and to which behaviour is intended to be generalised

(Araújo, et al., 2007; Pinder, Davids, et al., 2011b). The term representative design

captures the idea of sampling perceptual variables from an individual’s performance

environment to be designed into an experimental task (Brunswik, 1956). Recent

work has considered how the concept of representative design can be applied to the

study of sport performance (Araújo, et al., 2006; Araújo, et al., 2007). Inspired by

Brunswik’s (1956) insights, the term representative learning design (RLD) has been

proposed to highlight the importance of creating representative environments for

learning skills and developing expertise (Davids, et al., 2012; Pinder, Davids, et al.,

2011b).

Previous empirical work on RLD (Pinder, Davids, et al., 2011a) has focussed

on visual information provided during practice in training environments of elite

athlete programmes (Barris, Davids, et al., 2013), and changes to the complexity of


organisation in tasks for practising passing skills in team games (Travassos, et al.,

2012). These examples advocate expertise acquisition by nurturing the relationship

between key environmental information sources and coordination tendencies of a

performer in order for more adaptable and effective movement behaviours to emerge

(Davids, et al., 2013; Phillips, Davids, Renshaw, & Portus, 2010). From this

perspective the development of expertise is predicated on the accurate simulation of

key performance constraints during practice/learning. This approach differs from

traditional methods of decomposing tasks to isolate individual components, in order

to manage the information load confronting learners (Phillips, et al., 2010; Pinder,

Renshaw, & Davids, 2013).

An aspect of RLD that needs attention in future conceptualisation of learning

and practice is the role of affective constraints on behaviour (for initial discussions

see, Pinder, et al., 2014). In sport, performers need to be able to adapt to task

constraints while performing under differing emotional states induced in competitive

performance that can influence their cognitions, perceptions and actions (Jones,

2003; Lewis, 2004). Previous work investigating affect in sport performance has

tended to focus on capturing the emotions of athletes in ‘snapshots’ of performance

at one point in time, such as before or after competition (for a recent example see

Lane, et al., 2012). Such an approach, however, has not considered how emotions

might continuously interact with intentions, cognitions, perception and actions to

constrain the acquisition of functional coordination patterns and the development of

expertise. A holistic approach should consider task demands of learning

environments and the dynamic psychological state of each individual learner as

interacting constraints influencing behavioural (perception-action couplings),

cognitive, and emotional tendencies (Davids, et al., 2013; Newell, 1986). These


ideas suggest how sport psychologists and practitioners may seek ways to sample the

intensity of emotionally-charged performance conditions in learning environments

and practice simulations. To address this issue, the following sections of this paper,

will discuss why and how emotions could be incorporated into representative

learning designs to enhance acquisition of expertise in sport.

Affective learning design

Yet to be seen in the literature is a principled exploration of the role of

emotions in developing expertise in sport (Pinder, et al., 2014; Renshaw, Headrick,

& Davids, 2014). The role of affect in developing expertise might be harnessed by

adhering to two principles: (i) the design of emotion-laden learning experiences that

effectively simulate the constraints and demands of performance environments in

sport; (ii) recognising individualised emotional and behavioural tendencies that are

indicative of learning. These principles suggest, two complementary perspectives on

Affective Learning Design (ALD), linking the development of representative learning

designs with the identification and recognition of individual behavioural tendencies

exhibited while learning.

Benefits of creating emotion-laden learning events have been demonstrated

within the psychology literature. Emotions influence perceptions, actions and

intentions during decision-making, with the intensity of emotion generated reflecting

the significance of stimuli to an individual, shaping the strength of the response on

the visual cortex (Pessoa, 2011). Emotion also acts to strengthen memories (positive

or negative) and produces greater engagement in ambiguous, unpredictable, or

threatening situations when individual and group goals are influenced (e.g. learning

when failure might have significant consequences such as non-selection for a team,


or a team failing to qualify for a future event) (LaBar & Cabeza, 2006; Pessoa,

2011).

Despite these proposed benefits, the role of emotion in the pursuit of

expertise in sport has often been neglected (or removed) during practice because

emotion-laden responses are traditionally considered irrational or instinctive, and

therefore perceived as a negative influence on action (Hutto, 2012; Jarvilehto,

2000a). A neglected issue is that a significant constraint in competitive performance

environments is the emergent emotional tendencies of each individual. Therefore a

key question is, how can individuals be supported while exploring and exploiting

emotional constraints when learning to perform in competitive performance

environments? Emotionless responses made from a purely informational stance have

been described as ‘cold cognition’, and emotion-laden responses as ‘hot cognition’

(Abelson, 1963). The expression of ‘sit on your hands’ in relation to choosing a

move in a game of chess exemplifies a traditional view that it is necessary to

suppress or remove emotions in order to make more rational decisions (i.e. cold

cognition) (Charness, et al., 2004). However, during competitive performance in

sport, athletes are often not afforded this ‘thinking’ time and need to be able to act

immediately based on the initial, fleeting interaction between their perceptions of the

task and pre-existing physical, cognitive, and emotional capabilities (Davids, 2012).

This performance capacity has been referred to as ‘ultrafast’ behaviours (Riley,

Shockley, & Van Orden, 2012).

Progress in understanding emotions during learning has also been limited by

a tendency towards traditional linear thinking, where cognitions related to events are

considered to result in preconceived emotional reactions (Lewis & Granic, 2000).

Some psychologists have recently begun to acknowledge the advantages of


considering humans as complex, highly integrated dynamical systems in explaining

emergent behaviours (Lewis, 1996; Lewis & Granic, 2000). From this approach

cognition and emotion are considered to constrain each other interactively (similar to

processes of perception and action), with cognitions bringing about emotions, and

emotions shaping cognitions (Lewis, 2004). This cyclical interaction underpins the

emergent self-organisation of cognitions and emotions experienced during task

performance (Lewis, 1996, 2000a). Established emotional experiences represent

stable patterns of behaviour that are formed when emotional and cognitive

changes/responses become embodied in behavioural tendencies (Lewis, 2000a). In

other words, intertwined emotions, cognitions, and actions can become stable,

characteristic responses to particular experiences (Lewis, 1996, 2004). In this line of

thinking, affect, cognition, and behaviours exhibit self-organisational tendencies to

underpin characteristic performance responses, and shape the intrinsic dynamics of

an individual (Davids, et al., 2001; Schöner, Zanone, & Kelso, 1992). For example,

in the development of personality, trait-like behaviours, thoughts and feelings

become predictable, stable responses of an individual under certain performance

conditions (Lewis, 1996).

During the development of emotional interpretations, changes in performance

constraints may lead to metastable periods where an individual could rapidly transit

towards one of a ‘cluster’ of possible cognitive-emotive states (Hollis, et al., 2009;

Lewis, 2000b, 2004). When in a metastable performance region (for example during

learning), behavioural tendencies of an individual would be expected to fluctuate

(exhibit increased variability) until a more stable state of behaviour emerges (Chow,

et al., 2011; Hollis, et al., 2009). Accompanying this variability in performance

behaviours, variable and individualised emotional responses also emerge (Lewis,


1996, 2004). Much like movement variability, emotion during learning (and

performance) has previously been considered as ‘unwanted system noise’ (Davids, et

al., 2003; Smith & Thelen, 2003). An ecological dynamics approach questions this

assumption, suggesting that the presence of emotion during learning is indicative of a

performer being engaged in task performance as they seek to utilise available

affordances to satisfy their intentions and goals (Jones, 2003; Seifert, Button, et al.,

2013).

For example, gymnasts attempting routines on balance beams of increasing

height have been found to display performance decrements, elevated heart rate, and

increased prevalence of perceived dysfunctional emotions (e.g. reporting feeling

nervous or scared) particularly on a first attempt (Cottyn, et al., 2012). Similarly,

comparisons of performance during climbing traverses, identical in design but

differing in height from the ground, have revealed that higher traverses increased

anxiety, elevated heart rates, lengthened climbing duration, and increased

exploratory movements in climbers (Pijpers, et al., 2005). Such findings highlight

the intense emotions often involved with moving out of a ‘comfort zone’ when

confronted with a new or more challenging task. This idea can also be interpreted

through work in cybernetics where individuals are viewed to adapt to situations until

reaching a critical point where they must undergo a shift or reorganisation to

maintain effective action and emotion characteristics (see, Carver & Scheier, 1998,

2000; Carver, et al., 2000).

A relevant body of work has investigated the potential advantages to learning

outcomes when training under pressure and the constraints of induced performance

anxiety in a range of tasks (Oudejans, 2008; Oudejans & Nieuwenhuys, 2009;

Oudejans & Pijpers, 2009, 2010). This work focused on the task constraint of


anxiety and training under pressure, with findings providing clear implications for

developing context-specific expertise by acknowledging the role of emotions in

learning. For example, in a dart throwing task participants who trained under the

task constraint of mild anxiety were found to more successfully maintain their

performance levels in high anxiety conditions, compared with those who trained in

low anxiety conditions (Oudejans & Pijpers, 2010). In this case anxiety was

manipulated by positioning dart throwers at different heights on an indoor climbing

wall (also see, Oudejans & Pijpers, 2009). Similar findings were revealed in a study

comparing the role of pressure in a handgun shooting task involving police officers

(Oudejans, 2008). Here the control group (low pressure) shot at cardboard targets,

while a high pressure group shot at opponents who could fire back with marking

cartridges. Prior to practice, the performance of both groups was found to deteriorate

when switching from low to high pressure task constraints. After completing three

practice sessions, performance scores indicated that the shooting performance of the

experimental group was maintained for the high pressure condition. In comparison,

the performance of the control group deteriorated under high pressure as observed

prior to the practice sessions. Induced anxiety was again used as a task constraint

during practice sessions in an attempt to improve basketball free throw shooting

under pressure (Oudejans & Pijpers, 2009). Participants in an experimental group

were made aware that their practice sessions were being recorded, viewed and

evaluated, along with being constrained by simulated competitive performance

scenarios and the possibility of receiving performance rewards. As with the previous

examples, the experimental group was found to maintain free throw performance

during low pressure tasks into high pressure tasks. The performance of the control


group, who practiced under low anxiety, deteriorated in high pressure conditions

following five weeks of practice.

The findings of these studies have clear implications for how affective task

constraints can be manipulated for the acquisition of expertise in sport. The data

highlight that sport psychologists need to consider how behaviour and performance

outcomes can be constrained by simulated emotional and cognitive states of

individual performers during practice. In acquiring expertise, performers will

experience periods of failure or success as they strive to achieve a high level of

‘fitness’ for specific performance landscapes (Collins & MacNamara, 2012).

Learning environments need to be designed to include situation-specific

informational constraints that shape and regulate movement behaviours and the

emotional constraints of a task in relation to the intentions of a performer (Davids, et

al., 2001). From this approach, emotions are influenced by the constraints of the

task and also act as constraints on future behaviours emerging across interacting time

scales (i.e. performance, learning and development time scales) (Lewis, 2000a,

2004). Drawing on this interaction, ALD advocates for the design of emotion-laden

learning experiences that represent the constraints of competitive performance and

promote the acquisition of expertise within/for that context. Underpinning the design

of representative experiences is the observation and analysis of emotions in

conjunction with movement behaviour to identify periods of learning.

Affective learning design in practice

The intertwined relationship between movement behaviour and emotions

poses many challenges and implications for sport psychologists and other

practitioners interested in understanding how the concept of ALD can be applied to

the acquisition of expertise in sport. Key considerations for implementing ALD


include (i) adopting an individualised approach, (ii) acknowledging different time

scales of learning, and (iii) embedding emotions in situation -specific task

constraints. Sport psychologists implementing ALD need to sample, predict and plan

for the potential emotional and cognitive circumstances in competition, and

adequately sample them in learning simulations. This premise links to the two

previously identified principles of ALD regarding the design of representative

emotion-laden learning experiences, and identifying emotional and behavioural

tendencies that are indicative of learning. The following discussion of these ideas

includes a series of practical examples of how ALD might be embraced by sport

psychologists, pedagogues, coaches, and athletes.

The individualisation of affect

Of major significance for the design of affective learning environments is

catering for individual differences between performers. Sport psychologists must

collaborate with coaches to exploit their experiential knowledge to individually tailor

learning experiences based on skill level, personalities, learning styles, and

psychological strengths/weaknesses (Renshaw, Davids, Shuttleworth, & Chow,

2009). For example, it is worth considering some data on how skill-based

differences might interact with emotions to constrain cognitions, perceptions and

actions of different individuals (Seifert, Button & Davids, 2013). A comparison of

the performance of ice climbers revealed that the intra-individual movement choices

(e.g. kicking, hooking into the ice) and inter-limb coordination modes of novices

displayed less variability than those of experts (Seifert & Davids, 2012). In this

research novices tended to intentionally adopt an ‘X’ position with their arms and

legs that provided highly stable interactions with the surface of the ice. These

coordination patterns were functional for novices since they provided stability on the


ice surface. However, adoption of these highly secure patterns was not functional for

the goal of climbing the ice fall quickly, as demonstrated by the levels of variability

in positioning of the experts. The implication is that energy efficiency and

competitive performance were not prioritised in the goals of novice performers,

whose specific coordination tendencies emerged as a function of their uncertainty in

interacting with the ice surface.

In this example, the intentions (i.e. stable position vs. efficient and effective

climbing movements) of each performer, based on their individualised perception of

affordances, provide scope for a coach or sport psychologist to design targeted

learning events. Key constraints can then be implemented and manipulated to

simulate challenges that are anticipated to enhance situation-specific expertise at an

individualised level, based on identified stable emotional and behavioural tendencies.

For example a novice climber might be encouraged by a coach to move away from

the ‘X’ position through the narrowing of available climbing area, forcing the

individual to explore climbing options with foot and hand holds closer to the body.

In implementing this approach a coach could develop an understanding of the most

successful methods for pushing each performer into metastable regions, where

established action-emotion tendencies become destabilised. As a result, the

performer will be forced to explore performance environments simulated during

learning to harness new functional states of stable system organisation, or at least

experience situations with different task demands (Chow, et al., 2011; Renshaw, et

al., 2009). This approach is synonymous with psychological ‘profiling’ and shares

some ideas with the notion of individual zones of optimal functioning (IZOF) model

advocated by Hanin (e.g. Hanin, 2007b; Hanin & Hanina, 2009) in which the


interaction between emotions and actions during optimal performance is considered

to be highly individualised.

Time-scales and affects

The individualised nature of emotions must also take into account the

different interacting time scales of learning that influence the development of

expertise (Newell, et al., 2001). The critical relationship between the time scale of

perception and action (short term over seconds and minutes) and those of learning

and development (longer term over days, weeks and months) predicates how an

individual might approach specific situational constraints. From a complex systems

perspective, perception and action constrain the emergence of long term patterns or

behavioural states (Lewis, 2000a, 2002). Initial experiences of a performer will

influence how he/she approaches tasks in the future, which emphasises the

importance of tailoring the design of learning tasks to individual needs at all stages

of the expertise pathway (Côté, Baker, & Abernethy, 2003).

For example, qualitative evidence from interviews with expert team sport

athletes revealed that the roles and expectations of coaches (and performers) change

along the pathway to expert performance (Abernethy, Côté, & Baker, 2002).

Perceptions of ‘expert’ coaches at early stages of sport participation were based on

creating positive environments (leading to positive emotions) that were engaging and

fun while also developing basic skills. Essentially, these early experiences were more

concerned about meeting the basic psychological need of learners to demonstrate

competence (Renshaw, et al., 2012), leading to higher levels of intrinsic motivation

that sustain engagement over longer time frames necessary to achieve expertise. As

athletes progressed through the developmental phases (i.e. from Romance to

Precision to Integration, see Bloom, 1985) the relationship with the coach became


more tightly coupled and tended to increasingly emphasise the acquisition of sport

specific knowledge for managing the physical, emotional, and cognitive needs at an

individual level (Abernethy, et al., 2002; Côté, et al., 2003).

Hence, by designing learning environments that cater for changing emotions,

cognitions and actions, performers are more likely to engage with or ‘buy into’ the

rigorous demands of long term development programmes (Renshaw, Chow, et al.,

2010; Renshaw, et al., 2012). This reinforces the importance of recognising

individualised physical and psychological tendencies across various periods of

learning, as well as the critical role of a coach or sport psychologist in designing

learning programmes that simulate and sample the intended performance

environment to effectively accommodate such behavioural tendencies.

Emotions are embedded in situation-specific task constraints

Emotion-laden experiences are considered to energise behaviour and

facilitate an investment in tasks because emotions add context to actions, rather than

an athlete merely ‘going through the motions’ in isolated practice drills (Jones, 2003;

Renshaw, et al., 2012). Creating individual and/or group engagement in learning

experiences through the manipulation of specific constraints enhances the

representativeness of a practice task. Through the inclusion of situation-specific

information, the demands of a competitive performance environment can be

simulated (Pinder, Davids, et al., 2011b; Pinder, et al., 2014). Based on this premise,

to facilitate the holistic development of expertise, performers should be immersed in

learning environments that challenge and stimulate both physically and

psychologically to coincide with the constraints of prospective performance

environments (Davids, et al., 2013; Renshaw, et al., 2012).


For example, rather than allowing an athlete to practise shots on a driving

range, a coach might walk alongside a trainee golfer, creating specific ‘vignettes’

(e.g., 1 shot behind with one hole to play or 2 shots ahead in the same situation) to

simulate competitive performance conditions under which a learner might need to

adapt their golf shots (e.g., play more conservatively or take more risks). Some

previous work in team sports research has incorporated vignettes into the design of

practice and performance tasks to investigate how manipulating situational

constraints might influence emergent behaviours in athletes. In basketball 1v1 sub-

phases, the manipulation of instructional constraints to simulate competitive

performance conditions was found to influence the specific intentions and emergent

behaviours of attacking players (Cordovil, et al., 2009). In that case game time and

score-based scenarios were implemented to encourage players to experience adopting

risk-taking, conservative, or neutral strategies that might emerge in competitive game

play. Similarly, in football, 1v1 attacker-defender dyads, located at different

locations on the field of play (by manipulating distance to the goal area) were found

to constrain how attacking and defending players interacted with each other

(Headrick, Davids, et al., 2012). Providing contextual information through vignettes

engaged the players in the task by specifying goals or objectives that simulated

typical game situations for each field position.

Further work originating from elite sport programmes has discussed the

importance of designing practice tasks that effectively replicate competitive

performance conditions for athletes. An example is the development of the ‘Battle

Zone’ in cricket as an alternative to traditional, decomposed, net-based or centre

wicket batting and bowling practice (Renshaw, Chappell, Fitzgerald, & Davison,

2010). The Battle Zone concept combines a regulation cricket pitch with a


downscaled netted field area to increase involvement and intensity for all players,

compared with full sized centre wicket practice. Vignette-based tasks such as the

Battle Zone, maintain the critical performer-environment interactions while also

affording coaches the opportunity to manipulate specific performance constraints to

physically and psychologically engage batters, bowlers and fielders simultaneously

(Renshaw, Chappell, et al., 2010; Renshaw, et al., 2012).

Practice task designs, such as the Battle Zone, manipulate the space and time

demands on players which is captured by the Game Intensity Index (GII) concept

(Chow, Davids, Renshaw, & Button, 2013). The GII (pitch area in m2/number of

players) can be used in various team and invasion games to create game intensities

representative of competition, compare types of games, and cater for different levels

of expertise. Coaches can systematically manipulate GII to match task demands to

current performance capacity (i.e., place the player’s in their comfort zones) before

pushing learners into metastable regions that correspond with instability in

conjunction with increased range and intensity of emotions, cognitions and actions.

For example, if a coach wished to observe how a young player could cope at the next

performance level, (s)he could manipulate the GII to simulate the spatiotemporal

demands of that level.

In fast ball sports like cricket and baseball, the temporal demands of batting

become more severe as performance levels increase. In cricket, while present

methods of preparing for this added temporal constraint often include resorting to

bowling/ pitching machines or intensive net sessions with ‘throw-down’s by coaches

from shorter distances, previous research has shown that removing essential

information in practice tasks (i.e., the bowler) results in changes in batter’s timing

and co-ordination (Pinder, Davids, et al., 2011a; Renshaw, Oldham, Davids, &


Golds, 2007). A more effective way, could be to face current fast bowling teammates

in Battle Zone vignettes, to replicate the time demands of facing faster bowlers. For

example, to replicate a 150 kmh-1

delivery a 140 kmh-1

bowler would need to release

the ball closer (1.75 m) to the batter than the ‘legal’ delivery distance to replicate the

time (0.41 s) available when facing the 150 kmh-1

bowler. As well as requiring the

batter to adapt on a perception-action level, simulating the faster bowling speed also

enables the batter to experience the potentially intense emotions associated with

facing bowlers of this speed. These task constraints also allow learners to experience

the consequent changes in perception, cognitions and actions associated with the

interaction between internal and external constraints underpinning performance.

Other work in the sport of springboard diving studied the practice methods of

athletes from an elite-level squad (Barris, Davids, et al., 2013; Barris, Farrow, &

Davids, 2013). In these studies, elite divers were observed to baulk (preparation

occurs but divers do not leave the board) during practice when the preparation phase

was perceived as not being ideal for the performance of a selected dive (Barris,

Farrow, et al., 2013). This behaviour posed problems for performance in competitive

events where baulked dives result in reduced scores from judges. In a planned

intervention, divers were required to avoid baulking unless it was perceived that an

injury might occur. Barris and colleagues (2013) reported no significant differences

in movement patterns between baulked and completed dives under these new task

constraints. However, quantitative analyses of variability within conditions, revealed

greater consistency and lower levels of (dysfunctional) variability amongst dives

completed prior to the training program, and greater levels of (functional) variability

amongst dives completed after experiencing the training programme. It was

concluded that divers should be encouraged to complete (where safe) all attempts to


more functionally simulate the adaptive performance requirements of competition

conditions. From an ALD approach, data suggested that under these practice task

constraints, divers would be more frequently exposed to metastable regions enabling

them to explore variable take off positions. These metastable regions were expected

to enhance the development of expertise (through increased adaptability) by

encouraging divers to complete dives where less than ‘optimal’ preparatory

movements were evident. These changes to practice task design created more

physically and emotionally demanding performance environments that better

simulated competitive performance conditions. As predicted, the elite springboard

divers displayed greater consistency in key performance outcome (dive entry). At the

end of a twelve-week training program that required divers not to baulk, athletes

demonstrated enhanced performance through increased levels of functional

movement variability. Data suggested that the intervention resulted in them being

able to adapt their movements in the preparatory phase and complete good quality

dives under more varied take-off conditions. These results bring to light some

important practical implications for athletes in training and competition by means of

improving training representativeness, reducing performance anxiety and enhancing

feelings of self-confidence (Barris, Davids, et al., 2013).

Each of the examples in this section illustrate how including representative,

situation-specific constraints has the potential to embed emotions in learning

environments. Considering such examples, in conjunction with the previously

discussed body of work focussing on anxiety and training under pressure (e.g.

Oudejans, 2008; Oudejans & Nieuwenhuys, 2009; Oudejans & Pijpers, 2010),

provides support for embracing emotions present in learning environments. The

advantages to learning and performance outcomes reported (e.g. in dart throwing and


basketball shooting) when training with anxiety, and the well-established benefits of

creating representative learning environments provide diverse, yet complementary,

perspectives for how ALD can enhance the development of expertise in sport. By

implementing ideas such as these, sports psychologists and coaches will be able to

observe and analyse the integrated emotional and behavioural tendencies of athletes

during learning. In turn, the identification of these emergent physical and

psychological tendencies has the potential to underpin the design of further effective

learning experiences.

Conclusions

Founded on ecological dynamics principles, previous work has

conceptualised and advocated a representative learning design for effective

development of skill and expertise in sport. To take forward the understanding and

application of this approach the importance of emotions in learning has been

highlighted and introduced an integrated concept of ALD with potential scope for

future theoretical modelling. Two key interlinked principles of ALD have been

identified: (i) the design of emotion-laden learning experiences that effectively

simulate the constraints and demands of performance environments, and (ii),

recognising individualised emotional and behavioural tendencies that are associated

with different periods of learning. Here it has been argued that these key principles

of ALD will be valuable in the acquisition of sport expertise by considering affect,

cognitions, and actions together as intertwined individualised tendencies which

constrain performance and learning. Enhanced understanding of individualised

behavioural tendencies during learning will also aid the design of representative

learning environments that more effectively develop situation-specific skills.


The concept of ALD also advocates designing learning environments at an

individualised level, acknowledging different interacting time scales, and

implementing vignettes or scenarios to provide context to tasks. This allows

performers to experience the emotional feelings associated with performing in

learning situations that simulate the external task demands of a ‘new’ environment.

Therefore performers are provided the opportunity to experience how they would

(potentially) respond emotionally (e.g. know what emotions were created and how

intense they were), how this impacted on the way they thought (e.g. influencing their

intentions/goals/motivations), and acted (how this affected their actions). ALD also

allows the performer, sport psychologist, and coach to understand the impact of

being placed in a metastable region (i.e. in a learning task) and the influence this has

on affect, cognitions and behaviours. By recognising this interaction it is envisaged

that performers and sport psychologists will begin to understand that variability is a

normal (in fact desirable) consequence of learning that can be incorporated to

develop enhanced learning experiences in the future.

Future research should aim to investigate the relationship between affect,

cognition, and action during learning experiences to provide further support for this,

and potentially expanded, ALD models. Upholding a focus on individualised

approaches is imperative to effectively capture how individual learners interact with

specific task demands and environments. This theoretical conceptualisation of how

affect, cognition, and action interact provides implications for the design of

integrated, systems-oriented learning environments that enhance the acquisition of

expertise in sport through enhancing the functionality of individual-environment

relationships

Chapter 4: Development of a tool for monitoring emotions during learning in sport 67

Chapter 4: Development of a tool for monitoring emotions during

learning in sport

The Sport Learning and Emotions Questionnaire (SLEQ)

Following the conceptualisation of Affective Learning Design (ALD) in Stage

1, this chapter (Stage 2) discusses the development of a tool for measuring emotion

intensity during learning in sport. Through several phases of item refinement the

Sport Learning and Emotions Questionnaire (SLEQ) was conceived, providing a

suitable tool for tracking and measuring emotions as discussed during subsequent

stages of the PhD.

68 Chapter 4: Development of a tool for monitoring emotions during learning in sport

Overview

Measuring and analysing emotion during learning in sport is a complex and

challenging task that has often been overlooked or neglected. Existing methods tend

to be derived from tools that are developed in contexts other than sport, are designed

for pre / post competition contexts, focus on one or a select few emotions, and are

static rather than dynamic in nature. The following chapter takes on the challenge of

developing an emotions questionnaire designed within the context of learning in

sport, for use during learning in sport. Through several phases of inquiry, emotion

items (words / phrases) were identified, refined, compared, and validated. By means

of confirmatory factor analysis a four-factor, 17 item model emerged, forming the

structure of the Sport Learning and Emotions Questionnaire (SLEQ). The SLEQ

provides an overall score of emotion intensity, along with subscale scores of

Enjoyment, Nervousness, Fulfilment, and Anger. Designed to be implemented

across several interacting time scales, the SLEQ is an innovative and much needed

tool that has the potential to be adopted by researchers and sport practitioners

interested in tracking emotion intensity throughout various learning tasks.


Introduction

Research investigating the type, intensity, and influence of emotions

experienced in sport contexts is an area of great interest to athletes, coaches,

pedagogues, and sport psychologists alike. Previous work has discussed the role of

emotions in relation to: decision-making (Tenenbaum, Basevitch, Gershgoren, &

Filho, 2013); reaction time, force production and accuracy (Beatty, Fawver,

Hancock, & Janelle, 2014); interactions between dysfunctional and functional

emotions (Cottyn, et al., 2012); and goal orientations (Dewar & Kavussanu, 2012).

As such, emotions have the potential to interact with and influence physical,

cognitive, and motivational functioning of sport performers (Hanin, 2000b; Jones,

2003; Lazarus, 2000; Vallerand & Blanchard, 2000).

Understanding how emotions and actions might interact under the influence of

changing constraints provides many challenges regarding how to effectively measure

or quantify the emotions of an individual (Headrick, Renshaw, Davids, Pinder, &

Araújo, 2015). Typically, attempts to measure emotions involve the use of a self-

report tool or scale that is implemented before, during, or after an event or

performance (Hanin, 2000a; Jones, Lane, Bray, Uphill, & Catlin, 2005). Alternate

methods ask performers to recall emotions retrospectively in relation to specific

experiences such as best or worst performances (Cottyn, et al., 2012; Hanin, 2000a,

2007b; Tenenbaum & Elran, 2003). Other approaches have concentrated on single,

or a select few emotional constructs such as anxiety or happiness using specifically

designed measures (Nibbeling, Daanen, Gerritsma, Hofland, & Oudejans, 2012;

Pijpers, et al., 2003; Woodman, et al., 2009).

Currently employed techniques for measuring emotion can also be categorised

into individualised and group (standardised) approaches. The individualised


approach to identifying and measuring emotions in sport has been led by the work of

Hanin, culminating in the Individual Zones of Optimal Functioning (IZOF) model

(Hanin, 2000a, 2007a; Hanin & Syrjä, 1995). The IZOF model focuses on the

subjective emotional experiences of individual performers during successful, average

and poor performances. By interpreting the interaction between emotions (positive

vs. negative) and performance, predictions are made regarding the functionality of a

performer in specific situations (Hanin, 2000a). An IZOF orientated approach

affords many benefits for profiling the emotion-performance relationships of

individual performers, however, it is limited in terms of the extent to which data can

be compared with other studies, or athletes (Jones, et al., 2005).

Measuring emotions associated with performance from a group perspective

commonly involves the use of a standardised scale. Two popular scales or tools

adopted for use in sport are the Profile of Mood States (POMS, see McNair, Lorr, &

Droppleman, 1971), and the Positive and Negative Affect Schedule (PANAS, see

Watson, Clark, & Tellegen, 1988). While both of these tools (and versions of them)

have been used in sport environments they were originally designed for clinical or

everyday living contexts respectively, and are therefore not ideally suited to sport

populations (Jones, et al., 2005; Syrjä & Hanin, 1998). Other popular tools have

concentrated on measuring anxiety in relation to performance in isolation, for

example, the Competitive State Anxiety Inventory-2 (CSAI-2, see Martens, Burton,

Vealey, Bump, & Smith, 1990), and the Sport Anxiety Scale (SAS, see Smith, Smoll,

& Schutz, 1990). Tools such as these have been designed specifically for use in

sports contexts, but are limited due to the sole focus on single emotions and hence

there is a lack of consideration for other contributing factors (Hanin, 2007b; Jones, et


al., 2005). Such tools are also quite static in design, therefore not effectively

capturing the dynamic nature of emotions over different time scales.

Taking into account the advantages and disadvantages of the previous

approaches to studying emotions and performance, Jones and colleagues (2005)

developed a measure of precompetitive emotions, the Sport Emotion Questionnaire

(SEQ). The SEQ focuses on measuring the emotions experienced in relation to a

specific event or competition (validated for pre-competition). The rationale for this

design is to appropriately capture emotional experiences, which by nature are

associated with particular antecedents (e.g. an event, object, or person), are short

lived, and are often intense in comparison to other affective phenomena (Beedie,

Terry, & Lane, 2005; Lane & Terry, 2000; Vallerand & Blanchard, 2000; Watson &

Clark, 1994). Moreover, emotions are understood to have intentionality or be ‘about

something’ in particular (Solomon, 2008). This approach incorporates some context

to the SEQ rather than the use of ambiguous question stems that are open to

interpretation by respondents. For example the question stem of “how do you feel

right now?” could elicit responses relevant to the overall emotions of the respondent,

but irrelevant to a specific sport competition. The SEQ is based on five key

emotions (5-factor model) relevant to sport, with the final inventory containing

twenty-two items that have been found to be strongly linked to one of the factors.

The following sections provide a brief overview of the five key emotions; namely

anger, anxiety, dejection, excitement, happiness (see Jones, et al., 2005).

Anger, anxiety and dejection represent unpleasant experiences that are

experienced by sport performers, although the fact that they are categorised as

unpleasant does not mean they are always have a negative impact of performance, in

fact at certain intensities they may be neutral or even positive. Anger is an emotion


associated with arousal that is linked with an impulse to counterattack and engage in

aggressive behaviour in sport (Isberg, 2000; Lazarus, 2000). While often associated

with frustration and poor performance, anger can also be beneficial to performance

when directed at a person or object that has insulted or offended the individual

(Beedie, et al., 2005; Lane & Terry, 2000). Anxiety relates to apprehension or

uncertainty associated with an experience that is perceived as threatening to an

individual (Lazarus, 2000; Raglin & Hanin, 2000). Research on anxiety has

dominated the work on emotions in sport when compared with other emotions, and

has revealed that anxiety can be both beneficial and detrimental to performance

(Hanin, 2010; Jones, 1995). The third unpleasant emotion from the SEQ is dejection,

which has strong links with the more clinically orientated term of depression (Jones,

et al., 2005). Dejection is associated with feeling of deficiency and sorrow relating

to an individual’s perception of actual and expected progress for a task (Allen, Jones,

& Sheffield, 2009; Frijda, 1994).

Happiness and excitement are the key pleasant emotions included in the design

of the SEQ. Happiness is often used interchangeably with the more intense emotion

of joy, which is considered to be a core positive emotion (Jackson, 2000; Shaver,

Schwartz, Kirson, & O'Connor, 1987). The experience of happiness relates to an

individual making reasonable progress towards fulfilling, or having fulfilled a

desired goal. Excitement is a pleasant emotional state associated with heightened

arousal that is frequently referred to as ‘facilitative anxiety’, reflecting the

characteristics shared with the unpleasant emotion of anxiety (Jones, 1995). The

perception of and ability to cope with increased challenges, and meeting or

exceeding goals have been shown to be associated with excitement (Burton &

Naylor, 1997; Lazarus, 1991).


Whilst Jones and colleagues (2005) suggest that the SEQ can be modified for

use with regards to a current or upcoming competition, they do remind us that it is

only validated for pre-competition use. While a modified version of the SEQ may

provide some insights to emotional experiences during learning in sport, the key

limitation is that the items included in this inventory were developed and validated to

focus on competition contexts. Previous research has demonstrated key differences

between competition and learning contexts in terms of goal orientations, effort,

enjoyment, motivation, and performance outcomes (e.g. van de Pol & Kavussanu,

2011; van de Pol, et al., 2012a). However, despite these clear differences, the role of

emotions during learning in sport has so far received limited attention (Headrick, et

al., 2015). To address this issue there is a need for a measurement tool to be

developed that is dedicated and specific to assessing emotions experienced during

learning in sport. Such a tool would provide valuable information to athletes,

coaches, pedagogues, and practitioners by providing increased understanding of

emotions experienced alongside physical performance and skill development.

Development of the Sport Learning and Emotions Questionnaire (SLEQ)

The aim of the current study is to develop a field based tool that assesses

emotions experienced during learning events in sport (i.e. over the time scale of

learning). To maintain validity and sensitivity this tool is designed to exclusively

measure emotion through the subjective experiences of respondents. Therefore, the

SLEQ provides a method for establishing situation specific intensities of emotion,

considered to be one of the key dimensions of emotion (Jones, et al., 2005; Parrott,

2001). As such it is envisaged that the SLEQ will be suitable for use as a stand -

alone tool alongside reliable and specific measures of actions and cognitions to

provide a holistic and intertwined understanding of emergent behaviour.


The process for developing this tool will follow the general framework

outlined by Jones et al (2005) in development of the Sports Emotion Questionnaire

which was designed for use in relation to competition. The development of the

emotions tool will be presented in a series of sections representing individual phases

of the process. Therefore, each phase includes individual methods, results, and

discussion sub-sections. This will be followed by a general discussion section

summarising the overall development process and key implications regarding this

tool.

Chapter 4: Phase 1 – Identifying emotional items 75

Chapter 4: Phase 1 – Identifying emotional items

The aim of the first phase was to collect an inventory of words and phrases that

reflected the emotions experienced by individuals learning to perform a skill in sport.

The most frequently identified terms would then be compared with the items

identified in the SEQ of Jones et al. (2005) to determine whether learning and (pre)

competition contexts produce different responses. It was hypothesised that some key

items would be shared with the SEQ, but several unique items would also be

identified due to the emphasis on learning rather than competition contexts.

Methods

Participants

Participants were 126 male (n=62) and female (n=64) undergraduate students

studying exercise science and/or health and physical education at an Australian

university (mean age 19.68 ± 2.37 years). All participants reported previous and/or

current experiences of learning to perform skills in 25 different sports. Sports

participated in were: Football/Soccer (n=24), Netball (n=17), Touch Football (n=13),

Australian Rules Football (n=11), Athletics (n=8), Rugby Union (n=8), Volleyball

(n=6), Rugby League (n=5), Basketball (n=5), Swimming (n=4), Cricket (n=4),

Futsal (n=3), Dance (n=3), and others (n=15). Within these sports the highest levels

of representation reported by participants were: school (n=9), club (n=19); district

(n=14), regional (n=25), state (n=41), and national (n=19). Therefore this cohort of

participants represented a wide range of different sporting backgrounds, and

experience levels from which emotion items would be identified.

76 Chapter 4: Phase 1 – Identifying emotional items

Procedure

During scheduled class time, potential participants were briefed on the nature

of the project by a research team member and provided with ethical and informed

consent details. Consenting participants were asked to individually complete a brief

written survey comprising two key sections (see Appendix B). The first section

recorded anonymous participant information regarding age, gender, primary sport,

and highest level of representation (see participant information above). The second

section asked participants to identify and list words or phrases that reflected their

emotions when learning to perform a skill in sport, including periods of success and

failure. Participants were asked to spend at least 5 minutes completing the survey

individually. No context or examples were provided to participants in order to

remove any experimenter bias in responses.

Analysis

Survey responses were collated to record frequencies for each emotional word

or phrase. An exploratory principal component analysis with oblique rotation (direct

oblimin) was performed to determine how many of the identified emotion items

should be retained going forward. Initial analysis calculated Eigenvalues along with

percentages of total variance and cumulative variance for each of the identified items

(components). A Scree plot was also created to visually represent the importance of

each component. All analysis was performed with IBM SPSS Version 22 (IBM

SPSS Statistics for Windows, 2013).

Results & discussion

In response to the second section of the survey, participants recorded a total of

979 words or phrases describing their emotions when learning a skill in sport (Mean

responses per participant 7.77 ± 2.89). Within the 979 total responses, 204 unique


terms or phrases were reported. Using Kaiser’s criterion (extracts components with

eigenvalues > 1) 68 components were initially flagged for extraction representing

92.53% of cumulative variance. Visual assessment of the scree plot (see Figure 4.1)

was ambiguous for this number of components and therefore Kaiser’s criterion was

deemed inappropriate due to overestimation. Further inspection of eigenvalues and

total variance values revealed that 39 components produced eigenvalues above 2,

paired with a variance percentage above 1. More specifically, component 39

represented an eigenvalue of 2.050 compared with 1.998 for component 40, with

variance values of 1.01% and 0.98% respectively. The cumulative percentage of

variance at component 39 represented 70.85%, with components 40 - 204 each

representing less than 1.0% of total variance. The scree plot still remained largely

inconclusive at this number of components. Upon reflection of the emotional items

corresponding to these components it was deemed appropriate for the purposes of the

questionnaire to include components 1 to 39 in an initial list that could be refined

further in following phases (see Table 4.1).

In order to make early comparisons with the SEQ, the initial list of 39 items

was contrasted at face value with the preliminary 39 item SEQ (see Jones, et al.,

2005, p. 420). Comparing these items with the preliminary SEQ was appropriate

given that the preliminary SEQ items had not yet been subjected to factor analysis,

and subjective trimming to meet the requirements of a short questionnaire (final SEQ

includes 22 items only). Furthermore, by coincidence both the preliminary SEQ and

the initial list both included 39 items making for simple comparison. Contrasting the

two lists revealed that 17 items were common, however a further 22 items were

found to be unique to the phase 1 survey results. The clear discrepancy in items

identified suggests that there are differences in emotions between pre-competition


and learning contexts that warrant further exploration. Upon further consideration of

the 22 unique items, seven items were flagged as potentially inappropriate to carry

forward to further stages of questionnaire development. Two items (confident,

achievement) were identified as potentially representing cognitions rather than

emotions. The item ‘confident’ was removed from the list (as with Jones, et al.,

2005), however, it was decided to tentatively retain ‘achievement’ given that this

item was unique to the initial list and warranted further scrutiny. Three items

(exhausted, pain, tired) were removed because they related to physical states and/or

could be misinterpreted. A further item was considered too colloquial to be included

(stoked: an Australian slang term meaning 'excited' or 'happy'), and another (pride)

was combined with the more frequently identified term (proud). Therefore a total of

33 emotional items relating to learning remained to be subjected to further scrutiny in

the following phases.


Table 4.1. The list of 39 most identified emotional terms from the phase 1 survey. Items are split into

those that are common with the preliminary 39 item SEQ (Jones et al., 2005), and those unique to this

study.

Items in common with SEQ (17)

Angry Annoyed Anxious Content Disappointed

Energetic Excited Frustrated Fulfilled Happy

Joy Motivated Nervous Pressured Sad

Satisfied Stressed

Unique items (22)

Accomplishment Achievement Bored Competitive Confident*

Confusion Determined Eager Elated Embarrassed

Enjoyment Exhausted* Fear Fun Included

Pain* Pride* Proud Relieved Stoked*

Successful Tired*

*Indicates items removed (see text)


Figure 4.1. Scree plot displaying Eigenvalues for phase 1 items. Components to the left of the dashed vertical line were retained (1-39).


Following the identification and trimming of items, a pilot investigation was

completed to provide an initial practical comparison of SEQ and phase 1 items. This

pilot was carried out in an attempt to provide preliminary evidence highlighting the

irrelevance of some SEQ items to learning contexts. Therefore the purpose of this

exploratory study was to provide further (practical) support for the development of a

new learning specific questionnaire. Given that this pilot study was not integral to

the development of the questionnaire it was included in the appendices for reference

(see Appendix C, page 263).

Chapter 4: Phase 2 - Face validation of items 83

Chapter 4: Phase 2 - Face validation of items

Phase 2 was designed to assess the face validity of the 33 items identified and

retained from the initial phase 1 survey. This process involved verifying whether

each of the items were considered: (i) relevant to individual’s experiences during

learning in sport; and (ii) whether the same items related to any of the five key

emotions (anxiety, dejection, anger, excitement, happiness) comprising the five-

factor model of the SEQ (Jones, et al., 2005). It was anticipated that at least one of

the factors (most likely anxiety or dejection) would be found to be irrelevant due the

change in context between competition and learning.

Methods

Participants

Participants for this phase were 87 male (n=43) and female (n=44)

undergraduate students studying exercise science and/or health and physical

education at an Australian university (mean age 20.71 ± 3.85 years). No participants

from this phase of the study had previously completed the initial survey from phase

1, or the golf putting pilot. All participants reported previous and/or current

experiences of learning to perform skills spread across 23 different sports. These

sports included: Netball (n=12), Football/Soccer (n=11), Cricket (n=9), Basketball

(n=8), Touch Football (n=6), Rugby Union (n=5), Australian Football (n=4),

Swimming (n=4), Gymnastics (n=3), Volleyball (n=3), Tennis (n=3), Athletics

(n=3), Powerlifting (n=3), Water Polo (n=2), Rugby League (n=2), Martial Arts

(n=2), and other (n=7). While participating in these sports, the highest levels of

representation reported were: school (n=4), district (n=15), regional (n=20), state

(n=22), and national (n=7).

84 Chapter 4: Phase 2 - Face validation of items

Procedure

During regular class time (that was not related to the research theme),

prospective participants were briefed on the nature and requirements of the project

by a research team member and provided with ethical and informed consent details.

Consenting participants were asked to individually complete a brief written survey

(see Appendix D). As with phase 1 the first section recorded anonymous

background information regarding age, gender, primary sport, and highest level of

representation. The second section presented participants with a matrix where they

were asked to indicate (by ticking a box) whether each of the 33 items from phase 1

were (i) relevant to emotions they had experienced during learning in sport, and (ii)

related to any of the five key emotion factors from the SEQ (anxiety, dejection,

anger, excitement, happiness). Participants were asked to tick as many boxes as they

felt appropriate and leave all other boxes blank. Once again, no context or examples

were provided to participants in order to remove any leading bias in responses.

Analysis

Participant responses were collated to produce frequencies for items in terms of

overall relevance, and relevance to any of the five key emotions/factors. The

percentage of participants identifying each item as relevant was calculated and

reported in a summary table (Table 4.2) ranked by overall relevance.


Results derived from the percentage of participants indicating the relevance of

an item to their own experiences and the five key emotions are reported in Table 4.2.

These results reveal that all 33 items were considered relevant to some extent, with

26 items reported relevant by at least half of the participants. Seven items

(embarrassed, fear, sad, confusion, content, elated, bored) were deemed relevant by


less than 50% of participants. Furthermore, none of these seven items were related

to any of the five SEQ factors by more than 50% of participants.

Table 4.2 also reports the five items that were most frequently linked with each

of the five SEQ emotions/factors. This provided an opportunity to compare the items

associated with each emotion factor from the SEQ. The SEQ structure included five

items for both anxiety and dejection, along with four items each for anger,

excitement, and happiness (see Jones, et al., 2005, p. 431). Results indicate that

anxiety was most related to the items: anxious; nervous; pressure; stressed; fear.

With the exception of fear all of these items were identified by 50% or more

participants. Dejection was linked to: disappointment; embarrassed; sad; confusion;

annoyed. All of these items returned percentages of less than 50% suggesting that

participants did not highly relate the emotion of dejection with learning experiences.

Anger was related most frequently to: angry; frustrated; annoyed; disappointed;

stressed. However, only the first three items were identified as relevant to anger by

more than 50% of participants. More than 50% of participants linked excitement

with the items: excited; competitive; motivated; energetic; achievement. Happiness

was most related with: happy; achievement; enjoyment; accomplishment; fun. Each

of these items was identified by over 70% of participants as being related to

happiness with several further items considered relevant by over 50% of participants.

Interestingly the item ‘achievement’ was found to be the most relevant item

overall, with 50% of participants linking it to both excitement and happiness.

Therefore, despite original doubts surrounding its inclusion, these results make a

strong case for retaining this item as part of a provisional 33 item list.


Table 4.2. Percentage of participants reporting items to be relevant to learning a skill in sport, and to

any of the five key emotions. Items are presented in descending order of overall relevance.

Emotion Relev-

ance

Anxiety Dejec-

tion

Anger Excite-

ment

Happi-

ness

Achievement 95.40 6.90 3.45 3.45 55.17* 78.16*

Successful 91.95 4.60 1.15 2.30 48.28 72.41

Competitive 90.80 26.44 2.30 13.79 68.97* 13.79

Determined 90.80 12.64 2.30 6.90 48.28 31.03

Accomplishment 89.66 8.05 2.30 1.15 48.28 75.86*

Enjoyment 86.21 0 0 0 45.98 78.16*

Fun 85.06 0 0 0 43.68 74.71*

Motivated 85.06 5.75 0 3.45 59.77* 45.98

Happy 83.91 0 0 0 29.89 81.61*

Excited 81.61 2.30 0 0 78.16* 39.08

Energetic 77.01 2.30 0 3.45 57.47* 40.23

Nervous 75.86 65.52* 6.90 2.30 16.09 1.15

Pressure 75.86 58.62* 8.05 6.90 20.69 3.45

Frustrated 73.56 18.39 13.79 60.92* 1.15 1.15

Proud 73.56 3.45 0 0 27.59 64.37

Disappointed 70.11 16.09 35.63* 28.74* 4.60 0

Satisfied 68.97 2.30 2.30 0 26.44 60.92

Annoyed 66.67 10.34 14.94* 51.72* 0 0

Anxious 66.67 82.76* 4.60 1.15 10.34 0

Relieved 62.07 5.75 0 0 18.39 54.02

Eager 60.92 8.05 0 0 49.43 20.69

Included 60.92 4.60 1.15 0 24.14 51.72

Joy 59.77 0 0 0 31.03 56.32

Stressed 54.02 52.87* 4.60 17.24* 5.75 2.30

Fulfilled 52.87 0 0 0 17.24 48.28

Angry 50.57 3.45 8.05 74.71* 1.15 1.15

Embarrassed 44.83 29.89 24.14* 3.45 2.30 1.15

Fear 41.38 35.63* 9.20 4.60 4.60 0

Sad 41.38 17.24 24.14* 12.64 2.30 3.45

Confusion 37.93 17.24 16.09* 8.05 0 0

Content 37.93 1.15 2.30 0 9.20 33.33

Elated 29.89 0 0 0 20.69 29.89

Bored 21.84 2.30 14.94 3.45 0 0

* Indicates items are ranked in the top five for each emotion/factor (from SEQ) by percentage


Taking these results into consideration, it was concluded that the five key

emotions used as factors in the design of the SEQ were not appropriate to include on

a tool to record and measure emotions during learning in sport. On face value,

anxiety, excitement and happiness were found to be relevant with at least four items

linked to each by more than half of the participants. Only three items were linked to

anger by more than 50% of participants suggesting that this emotion/factor was

marginally relevant and potentially not warranting inclusion. No items were found to

be related to dejection by more than 50% of participants, with the item

‘Disappointed’ the highest at only 35.63%. In comparison, during the development

of the SEQ a total of 12 items were deemed relevant to dejection by more than 50%

of respondents. A further indication that the five-factor model from the SEQ may

not be appropriate was that several items (e.g. achievement, successful, motivated,

and disappointed) were found to be strongly linked (ranking in the top five by

percentage) at face value to two factors. This suggests that in the context of learning,

participants found it difficult to distinguish between the designated factors for

selected emotion items.

Overall relevance percentages for items that closely matched the five factors

also suggested that dejection was not relevant as a key emotion for learning. Results

from Table 4.2 reveal that anxiety (anxious – 66.67%); anger (angry – 50.57%);

excitement (excited – 81.61%); and happiness (happy – 83.91%) were all considered

relevant by more than half the participants. The fifth factor of dejection (or a

derivative of) was not listed as one of the 33 items in this phase, or in the initial list

of 204 items from phase 1, offering a possible explanation for why so few

participants found it relevant to any of the items. In a similar phase of their study

Jones et al. (2005) found that dejection and anger were also reported to be less


relevant, than anxiety, excitement and happiness. Reasons for this include that

dejection and anger are of negative valence, and may be experienced less frequently

though at a higher intensity (Diener, et al., 1985; Lane & Terry, 2000; Woodman, et

al., 2009). The predominance of items relating to happiness and excitement also

supports previous work suggesting that the study of positively valenced emotions is

essential for understanding sport contexts, as opposed to the traditional (clinical)

focus on negative emotions (Folkman, 2008; McCarthy, 2011; Skinner & Brewer,

2004).

Following the findings from phase 1 (and the golf putting pilot task) it was

established that the emotion items from the SEQ of Jones et al. (2005) were not all

appropriate for learning contexts. Phase 2 set out to establish the face validity of the

items and determine how relevant they are considered to the key factors/emotions

from the SEQ. Findings from this phase suggest that the factors from the SEQ were

not considered relevant to learning contexts. In particular the factor ‘dejection’ was

not linked closely with any of the items identified in phase 1. Therefore, the case for

designing a new tool that measures intensity of emotions during learning is

strengthened. This tool will include both emotion items and factors specific to the

context of learning in sport.

Chapter 4: Phase 3 – Assessment of item and factor structure 89

Chapter 4: Phase 3 – Assessment of item and factor structure

The preceding development phases identified the need for a specific tool

relating to emotions experienced during learning. The eventual tool is intended for

use during field-based learning / skill acquisition sessions. Therefore, this phase of

development is focussed on establishing factors (subscales) for the questionnaire and

validating the items with participants across a variety of sports. Based on the

responses from participants in this phase, the items will be refined, and factors will

be established to determine the design and structure of the final questionnaire.

Building on previous phases, it was hypothesised that positively valenced items

would be found most relevant to learning contexts. In terms of factor identification,

it was expected that four or five factors would emerge representing the key positive

and negative emotions reported. In order to develop a succinct, and easy to complete

questionnaire for use during applied learning sessions, it was also desirable that

approximately 10 items be trimmed from the list of 33 items coming into this phase

(SEQ includes 22 items). These factors, and associated loading items, were expected

to substantially differ from those developed in the SEQ.

An important decision within this phase of the questionnaire development

involved determining the number of factors, and number of individual items to link

to each factor. The choice of how many factors would make up the questionnaire

was largely dependent on the outcome of the factor analyses described in the

following sections. Due to the key similarities with the design of the SEQ it was

expected that four or five factors would emerge. At least two of these factors were

anticipated to represent positively valenced items, given the dominance of positive

items in the previous phases. Despite using the SEQ as a guide, this phase included a

90 Chapter 4: Phase 3 – Assessment of item and factor structure

key feature of questionnaire development that differed from that of the SEQ. The

key difference here is that Jones et al. (2005) designated the five factor model of the

SEQ prior to all subsequent phases of questionnaire development. Therefore, items

identified and refined through the various phases were modelled in relation to the

five targeted factors. The approach adopted for the development of this current

questionnaire allowed factors to emerge based on the results of item identification

and refinement in preceding phases of inquiry. As a consequence, selected emotion

items informed the identification of potential factors, rather than fitting items to

predetermined factors as in the SEQ. This process involved various stages of

statistical analysis paired with the experiential knowledge and interpretations of the

research team.

Following previous investigations and suggestions, the aim was to include

approximately four items in each of the identified factors. As mentioned previously,

a five factor model was used in the final version of the SEQ. Two of these factors

(anxiety, dejection) included 5 items, and the remaining three (excitement, anger,

happiness) 4 items (Jones, et al., 2005). In order to maintain internal consistency and

reliability of factors in brief questionnaires, it has been suggested that 4 items per

factor is ideal (Jackson & Marsh, 1996; Watson & Clark, 1997; Watson, et al., 1988).

Along the same lines, Bollen (1989) determined that factors should have at least 3

items or indicators with ‘nonzero loadings’. Based on these guidelines, the following

sections set out to detail the process of (i) identifying 4-5 possible factors that

represent the majority of responses through exploratory factor analysis (EFA); (ii)

refining and trimming the items that load onto these factors to ensure the brief nature

of the questionnaire; and (iii) verifying these factors and items that adequately load

onto each factor through confirmatory factor analysis (CFA).


Methods

Participants

Individual and groups of participants were recruited from sport academies,

institutes, clubs, and university exercise science / physical education courses. In total

342 male (n=180) and female (n=162) participants (mean age= 20.19 ± 3.62 years)

completed this phase of the questionnaire development, and were not involved in any

other phase. All participations reported involvement in competitive sport including:

Football/Soccer (n=50), Australian Football (n=40), Netball (n=28), Water Polo

(n=21), Athletics (n=15), Gymnastics (n=13), Touch Football (n=13), Rugby Union

(n=12), Rugby League (n=11), Swimming (n=11), Cricket (n=9), Basketball (n=9),

Powerlifting (n=6), Field Hockey (n=6), Volleyball (n=5), and others (n=93). Within

these sports the highest levels of participation were: school (n=6), club (n=65),

district (n=27), regional (n=47), state (n=107), and national (n=90). Length of

involvement in these sports was reported to range from 0.5 to 21 years (mean

duration= 8.83 ± 4.06 years). Following approval from a university ethics

committee, prospective participants were provided with informed consent details

prior to agreeing to complete the survey.

Procedure

Participants were asked to complete a two page survey based on their

experiences during a learning/skills focussed session. This survey (see Appendix E)

comprised of a section asking for anonymous background information including age,

gender, and sport background. The second section asked participants to read a list of

33 emotion items and rate them on a scale using the stem: ‘how you feel in relation

to the session you have just completed’. The 5 option response scale was previously

employed on the SEQ (Jones, et al., 2005) and POMS (McNair, et al., 1971).

Participants were asked to individually consider each item and then circle a number


from 0 – 4 based on how they felt in regards to the item during the session (0 = ‘not

at all’; 1 = ‘a little’; 2 = ‘moderately’; 3 = ‘quite a bit’; 4 = ‘extremely’). In

consultation with the coach, manager, or teacher / lecturer a member of the research

team attended identified sessions to confirm the focus on learning and/or skill

development over fitness, strength, conditioning, or game play. The research team

member took on a purely observational role during the session to ensure no

disruption was created to the normal training environment.

Analysis

Participant responses were initially collated in order to calculate and visually

inspect the basic frequency, rating, and mean values for each of the 33 items. The 33

items were then subjected to an exploratory factor analysis (EFA) using the principal

axis factor extraction method with oblique rotation (direct oblim – factors can

correlate, see Field, 2013; Williams, Onsman, & Brown, 2010), performed with

SPSS (IBM SPSS Statistics for Windows, 2013). The Kaiser-Meyer-Olkin (KMO)

measure was used to assess the sampling adequacy of the factor analysis. Following

the identification of the emergent factors, the groups of loading items were assessed

and interpreted to develop assumptions about what each factor related to.

Cronbach’s alpha (α) was also calculated for each extracted factor to assess the

reliability of the potential subscales when considering items loading above a value of

0.4. Reliability was also assessed following refinement and trimming of items to suit

the brief nature of the proposed questionnaire.

Using the identified factors as latent variables, the proposed questionnaire

design was subjected to confirmatory factor analysis (CFA) using structural equation

modelling (SEM) techniques. Factors were distinguished using the maximum

likelihood estimation method and were permitted to intercorrelate freely. To assess


the normality (multivariate kurtosis) of data the Critical Ratio (C.R) was inspected

(Byrne, 2010). Factor loadings were taken into account to assess model design,

following loading values (i.e. factor loadings > 0.6) from the SEQ (Field, 2013;

Jones, et al., 2005). Modifications to the model were also identified and subjectively

interpreted case-by-case based on the value of Modification Indices (MI), and

whether the suggested changes made logical sense (i.e. error covariances between

related items) within the model design. In the event of unacceptable model fit,

modifications were made one at a time to independently assess the impact of the

change on model fit criteria (Byrne, 2010). Goodness-of-fit of the model was

evaluated through the overall chi square (χ2) value (with χ

2/df ratio), along with a two

index method comprising the comparative fit index (CFI - Bentler, 1990) and root

mean square error approximation (RMSEA - Steiger & Lind, 1980). The use of chi

square, CFI and RMSEA (with 90% confidence intervals) values as methods for

assessing model fit have been extensively advocated previously (Bentler, 2007;

Browne & Cudeck, 1993; Byrne, 2010; Hu & Bentler, 1999; Jackson, Gillaspy, &

Purc-Stephenson, 2009). This phase of analysis was performed using SPSS AMOS

(IBM SPSS AMOS Statistics for Windows, 2013).

Following the identification of a suitable model correlation analyses were

performed on the mean scores for each factor to determine relationships between the

proposed subscales. A one-way repeated measures ANOVA with pairwise

comparisons, Bonferroni corrections, and Greenhouse-Geisser corrections to

sphericity was also performed to assess statistical differences between factor scores.



Descriptive results

A summary of results for all 33 items is presented in Table 4.3 detailing the

overall mean ratings, along with a breakdown of rating frequencies for each item.

Interesting to note is the trend of positively valenced items generally rating at higher

scores than negatively valances items. This supports the findings from the golf

putting pilot investigation where positive items were more predominant overall.


Table 4.3. Summary of survey results presented in descending order of mean rating.

Item Mean rating

± SD # 0 # 1 # 2 # 3 # 4

% yes

(i.e. 1-4)

Motivated 3.23 ± 0.87 5 8 45 130 154 98.54

Determined 3.20 ± 0.88 5 11 43 136 147 98.54

Enjoyment 3.16 ± 0.95 7 15 44 125 151 97.95

Included 3.07 ± 0.94 6 13 64 126 133 98.25

Fun 3.05 ± 1.01 7 25 48 125 137 97.95

Happy 3.00 ± 0.90 4 15 68 144 111 98.83

Excited 2.91 ± 1.07 10 29 65 116 122 97.08

Competitive 2.86 ± 1.09 15 21 75 116 115 95.61

Accomplishment 2.86 ± 0.82 5 12 75 184 66 98.54

Successful 2.84 ± 0.93 6 18 88 142 88 98.25

Energetic 2.82 ± 0.98 8 23 82 138 91 97.66

Achievement 2.81 ± 0.84 5 13 90 169 65 98.54

Eager 2.79 ± 1.02 9 29 80 132 92 97.37

Joy 2.78 ± 1.02 12 23 82 136 89 96.49

Satisfied 2.58 ± 1.06 15 38 90 131 68 95.61

Fulfilled 2.50 ± 1.03 14 39 106 127 56 95.91

Proud 2.50 ± 1.06 16 45 91 133 57 95.32

Content 2.21 ± 1.08 29 48 124 104 37 91.52

Elated 1.94 ± 1.17 53 54 129 73 33 84.50

Relieved 1.91 ± 1.13 46 69 124 76 27 86.55

Pressure 1.58 ± 1.18 84 73 105 64 16 75.44

Nervous 1.28 ± 1.19 112 99 71 43 17 67.25

Frustrated 1.26 ± 1.16 115 95 73 47 12 66.37

Stressed 1.09 ± 1.10 137 85 81 31 8 59.94

Anxious 1.07 ± 1.11 140 88 72 33 9 59.06

Annoyed 1.00 ± 1.01 135 108 65 31 3 60.53

Disappointed 0.95 ± 1.04 148 103 60 23 8 56.73

Angry 0.93 ± 1.01 148 104 61 24 5 56.73

Fear 0.83 ± 1.02 173 89 50 26 4 49.42

Confusion 0.75 ± 0.88 165 115 45 16 1 51.75

Embarrassed 0.68 ± 0.88 188 94 43 17 0 45.03

Bored 0.62 ± 0.99 217 72 29 15 9 36.55

Sad 0.40 ± 0.73 243 69 25 2 3 28.95


Exploratory factor analysis

Sampling adequacy was found to be highly acceptable (KMO = .90).

Furthermore the KMO values for individual items were also highly acceptable with

values ranging from .76 to .95 (values over .50 considered acceptable, Field, 2013).

The EFA of potential factors and loading items initially extracted six factors with

Eigenvalues over Kaiser’s criterion (1), representing 63.92% of total variance.

However, six factors were not deemed appropriate for the purposes of the eventual

questionnaire due to several of these factors having less than 3 items loading above

an acceptable value of 0.4 (Field, 2013), and the items loading to each factor not

displaying clear trends or themes. Similar issues with loading items arose for the

extraction of five factors, along with ambiguous results from the scree plot in both

cases. Extracting four factors revealed more promising factor/item groupings, with

similar items tending to load on to common factors. These four factors all had

eigenvalues above 1 and together represented 56.78% of sample variance. Rotated

loadings for each item on the four extracted factors are presented in Table 4.9. Based

on the nature of the items strongly loading (i.e. above +/- 0.4) to each of the extracted

factors, it was deduced that the factors could tentatively be labelled as ‘Enjoyment’,

‘Nervousness’, ‘Satisfaction’, and ‘Annoyance’. The four labels were indicative of

the potential subscales and structure of the eventual Sport Learning and Emotions

Questionnaire (SLEQ). Reliability of items relating to the proposed subscales was

found to be high with each factor returning a Cronbach’s α value of at least .85 (see

Table 4.4).


Table 4.4. Item factor loadings, eigenvalues, variance %, and Cronbach’s α for each of the four

extracted factors and/or questionnaire subscales.

Item Factor 1 Factor 2 Factor 3 Factor 4

Accomplishment .071 -.037 .685 .077

Achievement .047 -.043 .674 .052

Angry -.011 -.005 -.014 .795

Annoyed .007 -.106 -.103 .911

Anxious -.013 .562 -.093 .221

Bored -.587 .027 .165 .282

Competitive .473 .067 .111 .125

Confusion -.060 .199 .126 .314

Content .020 -.052 .437 -.088

Determined .423 .020 .282 .086

Disappointed .013 .256 -.113 .540

Eager .595 .070 .018 .031

Elated .139 -.024 .419 .094

Embarrassed -.054 .275 .077 .399

Energetic .565 -.067 .228 .073

Enjoyment .863 -.014 -.024 .023

Excited .782 .115 .040 .033

Fear .006 .692 -.005 .059

Frustrated .028 .167 -.087 .664

Fulfilled .214 -.032 .605 .001

Fun .819 .028 .041 -.007

Happy .779 -.015 .070 -.041

Included .669 -.079 .015 -.058

Joy .782 -.069 .090 .020

Motivated .594 .059 .185 .024

Nervous .077 .970 -.017 -.153

Pressure .073 .737 -.015 .015

Proud .174 .059 .583 -.035

Relieved -.060 .188 .507 -.097

Sad -.215 .393 .153 .246

Satisfied .038 -.055 .743 -.062

Stressed -.006 .715 .025 .116

Successful .022 -.028 .770 -.016

Eigenvalues 9.48 6.05 1.77 1.43

Variance % 28.72 18.35 5.37 4.34

Cronbach’s α .93 .87 .86 .85

Note: Cronbach’s α calculated using items with factor loadings above 0.4 only (in bold)


Following the initial identification of factors, the items loading onto each were

subjected to further scrutiny in order to meet the goal of including at least 4 items per

factor while also ensuring a brief (approximately 20 item) questionnaire. Factor 1

(Enjoyment) had 11 items loading at values of 0.4 or higher and therefore substantial

refinement was required. Closer examination of factor loadings revealed a

substantial drop in values between the fifth (happy .779), and sixth (included .669)

items, supporting the retention of the top 5 items. These items all corresponded well

with each other in terms of reflecting the inherent happiness and enjoyment involved

with learning experiences. The items deemed not suitable on account of factor

loadings (included, eager, motivated, energetic, competitive, determined) were also

deemed to be not as relevant to the theme of enjoyment. On a side note, the item

‘bored’ also loaded highly to factor 1, however, as a negative value. The decision

was made not to retain this item given it corresponded with a weaker factor loading

than the top 5 items and also did not correspond with the theme of enjoyment.

Factor 2 (Nervousness) included 5 items that loaded over the value of 0.4. The

fifth item ‘anxious’ was flagged as a potentially weak item given the decline in factor

loading between it and the other four items. Therefore, all five items were retained at

this stage, with the possibility of trimming ‘anxious’ in subsequent analysis. The

third factor (Satisfaction) was also tentatively trimmed to five items from the initial

nine items loading above 0.4. This decision was based on a combination of factor

loadings and results from earlier phases. The items ‘elated’ and ‘content’ were

removed on account of the comparatively weak factor loadings in relation to the top

five items. The item ‘relieved’ was removed due to the drop in factor loading

between it and the top four items in particular. The decision not to retain ‘proud’

was a difficult one; however, 6 items were not deemed suitable for a single factor


(Jackson & Marsh, 1996; Watson & Clark, 1997; Watson, et al., 1988).

Furthermore, results from previous phases indicated that ‘proud’ was not deemed as

relevant to learning by participants in comparison to the higher ranking items. The

fifth item ‘fulfilled’ was also flagged for potential removal at later stages due the

difference in factor loading between it and the top four items. Factor 4 (Annoyance)

was a more straightforward decision with only 4 items loading to this factor by more

than 0.4. The fifth highest loading item (embarrassed) was considered for retention,

however, it also loaded in the top seven items for the second factor. Therefore, it

was deemed inappropriate to include ‘embarrassed’ on either factor. Embarrassed

was also considered to not fit well with the overriding theme of the four retained

items. Furthermore, ‘disappointed’ was flagged as a potentially weak item that

might come under further scrutiny.

Reviewing these refinements to the tentative questionnaire design reveals a

four factor structure with sets of five items loading on three of the factors, and four

items loading on the remaining factor. In comparison, the final 22 item version of

the SEQ included five factors, two with five loading items, and three with four

loading items. The tentative 19 item design for the SLEQ retains reliability for the

four proposed subscales with alpha (α) values of at least .85, as reported in Table 4.5.

Item origin results show that the majority of items (13 of 19) were present in both the

phase 1 list (Table 4.1) and the preliminary SEQ of Jones et al. (2005). However,

five of the items shared with the preliminary SEQ (pressure, stressed, satisfied,

fulfilled, frustrated) were not retained in the final version of the SEQ. Factor 3

(Satisfaction) included the highest proportion of items exclusive to the phase 1 list (3

of 5) suggesting that feelings of achievement and success relate highly with

participants in learning contexts. Furthermore, neither of the factor 3 items included


in both the preliminary SEQ and phase 1 list (satisfied, fulfilled) were retained in the

final version of the SEQ. This effectively means that the items that making up factor

3 are unique to this learning focussed emotion questionnaire. These findings once

again illustrate the difference between competition and learning contexts in terms of

participant experiences.

Table 4. 5. Proposed factor and item structure following EFA. Item origin: Phase 1: item originates

from phase 1 list only; Both: item originates from the preliminary SEQ and Phase 1.

Potential

Factors Items

Factor

Loadings

Cronbach’s

alpha (α)

Item

Origin

Factor 1

Enjoyment .863

.93

Phase 1

Fun .819 Phase 1

Excited .782 Both

Joy .782 Both

Happy .779 Both

Factor 2

Nervous .970

.87

Both

Pressure .737 Both*

Stressed .715 Both*

Fear .692 Phase 1

Anxious .562 Both

Factor 3

Successful .770

.86

Phase 1

Satisfied .743 Both*

Accomplishment .685 Phase 1

Achievement .674 Phase 1

Fulfilled .605 Both*

Factor 4

Annoyed .911

.85

Both

Angry .795 Both

Frustrated .664 Both*

Disappointed .540 Both

*item not included on the final version of the SEQ


Confirmatory factor analysis

Model 1

Following the development of the tentative four factor, nineteen item model

that emerged from the EFA, the next step was to assess the suitability of this model

structure through CFA. The proposed model (see Figure 4.2) returned a Chi-square

(χ2) value of 564.915, with 146 degrees of freedom, and a probability level of .000.

The χ2/ df ratio (3.87) exceeded the suggested cut-off value of ≤ 3 (Schreiber, Nora,

Stage, Barlow, & King, 2006). The relatively high χ2

value and significant

probability value suggest a poor model fit, however, the results also revealed a high

critical ratio (C.R.= 24.075) indicating multivariate nonnormality (kurtosis) in the

data sample (Byrne, 2010). Chi-square values have been suggested to be highly

sensitive to sample size and nonormality of data (high C.R. value), which can

produce inflated χ2

values and misleading significant probability levels (Curran,

West, & Finch, 1996; Steiger, 1990). No evidence of multivariate outliers was

observed for this model.

Continuing with the analysis of goodness-of-fit, the model initially produced

unacceptable CFI (.899), and RMSEA (.092 [.084, .100]

) values. Suggested cut-offs for

these goodness-of-fit indicators are CFI ≥.95 and RMSEA <.06 (Hu & Bentler, 1999;

Schreiber, et al., 2006). Upon examination of suggested modification indices it was

observed that there was some potential to improve model fit by modifying error

covariances. Potential modifications were considered suitable if MI values for

pairings were substantially higher than other pairings, and were appropriate in terms

of the nature of the paired items (i.e. items of similar meaning, see Byrne, 2010).

The most obvious pairing of error covariances for modification was those associated

with items 14 and 15 from factor 3. By adding this error covariance (MI = 110.339)


the modified model showed some evidence of improved fit (χ2

= 443.033; CFI =

.913; RMSEA = .087 [.078, .097]

). Two further rounds of modifications to error

covariances also produced slight improvement to model fit as follows: items 18-19

(MI = 36.589; χ2

= 400.471; CFI = .925; RMSEA = .066 [.058, .075]

), and items 1-5 (MI

= 11.199; χ2

= 357.113; CFI = .948; RMSEA = .081 [.072, .092]

). Despite the addition

of these three modifications the originally proposed factor-item structure did not

reach an acceptable level of fit. Therefore, it was concluded that this particular

model structure was not appropriate.


Figure 4.2. Model 1 with modifications. Values on model represent (left-to-right) error covariances,

standardised factor loadings, and factor correlations.

(χ2

= 357.113; CFI = .948; RMSEA = .081 [.072, .092]

)


Model 2

Based on the factor loadings from the EFA stage and observations from model

1 the next step was to investigate removing the items flagged as potentially weak (i.e.

disappointed, anxious, fulfilled). Both ‘disappointed’ and ‘anxious’ loaded at values

less than 0.6 to their respective factors in the EFA stage. Furthermore, throughout

the rounds of modification both these items continued to be the weakest loading

items on the relevant factors (particularly disappointed). On the other hand,

throughout the rounds of modification ‘fulfilled’ maintained a factor loading value

on par with the other four items loading on factor 3. As a result it was decided that a

revised model would be analysed with ‘disappointed’ and ‘anxious’ no longer

considered as loading items. The final version of the SEQ also included items with

factor loadings of at least 0.6, further justifying the removal of these two weaker

items. Therefore, model 2 was again based on a four-factor structure with five items

loading on factors 1 and 3, four items on factor 2, and three items on factor 4.

Initial inspections of model 2 (Figure 4.3) suggested an enhanced model fit as

opposed to model 1 (χ2

= 407.369; CFI = .920; RMSEA = .087 [.078, .097]

). Probability

remained the same (.000), while both the degrees of freedom (113) and χ2/ df ratio

(3.61) decreased as expected with fewer items. The C.R remained high (22.020) and

once again no outliers were observed. Modification indices suggested that error

covariances for items 13-14 would improve fit (MI = 110.297). This modification

substantially improved the model fit (χ2

= 270.283; CFI = .957; RMSEA = .064 [.055,

.074]), to a point where CFI was acceptable. Further appropriate modifications to error

covariances on items 1-5 (MI = 15.484; χ2

= 253.532; CFI = .961; RMSEA = .061

[.051, .071]), and 1-4 (MI = 11.998; χ

2 = 238.682; CFI = .965; RMSEA = .059

[.048, .069])

resulted in more than acceptable values for CFI, RMSEA, and χ2/ df ratio (2.17).


Figure 4.3. Model 2 with modifications. Values represent (left-to-right) error covariances,

standardised factor loadings, and factor correlations.

(χ2

= 238.682; CFI = .965; RMSEA = .059 [.048, .069]

)


Modification to error covariances were deemed suitable given the MI values of

pairings in each round, the similarity of these items, and that both items of each

pairing loaded on the same factor. An equivalent process was followed to determine

whether also removing ‘fulfilled’ would produce further enhanced model fit.

Following two rounds of suitable modification to error covariances values of cut-off

indices were χ2

(253.450), CFI (.954), and RMSEA (.069 [.059, .080]

) indicating that the

removal of this item did not improve model fit. The fit of the model when only

removing ‘disappointed’, the lowest loading item from EFA and model 1 was also

investigated. Following three suitable modifications to error covariances this model

structure again did not suggest a better fit than that of model 2 (χ2

= 303.660; CFI =

.955; RMSEA = .064 [.055, .074]

).

Results for the unstandardized estimates revealed that all were statistically

significant with individual item C.R values meeting the cut-off value of >1.96

(Arbuckle, 2013; Byrne, 2010). Standardised estimates (factor loadings) and error

variances are reported on Table 4.6 for the factor-item structure represented by

model 2. All items loaded on to the relevant factors with values ranging from .622 -

.915 (mean = .787). These values were similar, if not stronger to those observed in

the final version of the SEQ which ranged from .603 - .820 (mean = .747).

Reliability for the revised model structure was slightly reduced for factors/subscales

2 and 4 (See Table 4.6), however, all values remained strong and comparable with

those of the SEQ (Jones, et al., 2005).


Table 4.6. Factor loadings, error variances, and subscale reliability (α) for model 2 following CFA.

Item origin: Phase 1: item originates from phase 1 list only; Both: item originates from the

preliminary SEQ and Phase 1.

Factors Items Factor

Loadings

Error

Variance

Cronbach’s

Alpha (α)

Item

Origin

Factor 1

Happy .902 .150 Both

Fun .884 .222 Phase 1

Joy .879 .235 .93 Both

Enjoyment .770 .368 Phase 1

Excited .765 .474 Both

Factor 2

Nervous .859 .369

.86

Both

Stressed .779 .474 Both*

Pressure .776 .554 Both*

Fear .714 .506 Phase 1

Factor 3

Satisfied .784 .434 Both*

Fulfilled .781 .411 Both*

Successful .766 .354 .86 Phase 1

Accomplishment .659 .376 Phase 1

Achievement .622 .429 Phase 1

Factor 4

Annoyed .915 .167 Both

Angry .848 .286 .84 Both

Frustrated .681 .724 Both*

*item not included on the final version of the SEQ

Table 4.7. Mean score and correlations for each factor / subscale.

Factor / Subscale Mean score ± SD Factor 1 Factor 2 Factor 3 Factor 4

Factor 1 2.98 ± .87 1

Factor 2 1.19 ± .94 .02 1

Factor 3 2.72 ± .75 .62** .12* 1

Factor 4 1.06 ± .93 -.17** .49** -.06 1

Total SLEQ score 7.95 ± 2.14

** Significant correlation (p < .01), * Significant correlation (p < .05)


Sport learning and emotion questionnaire design

On account of the acceptable goodness-of-fit values, and strong (> 0.6) factor

loadings it was concluded that removing the items ‘disappointed’ and ‘anxious’ from

the model design was correct. Therefore, the design of the SLEQ would be

underpinned by the structure of model 2, with four factors comprising of five, four,

five, and three items respectively (17 items in total). This factor-item structure met

the guidelines discussed earlier in terms of addressing internal consistency and

representing a particular factor adequately (Bollen, 1989; Jackson & Marsh, 1996).

Factor 4, with three loading items, only just met these suggestions, however, as

established earlier, including the fourth item was found to diminish the goodness-of-

fit of the overall model.

Turning to the final questionnaire design, each of the four factors represents a

subscale for which the questionnaire scoring system would be based. Following the

simple scoring system from the SEQ (Jones, et al., 2005), the sum of item ratings

(i.e. 0-4) in each subscale is divided by the number items in each subscale to produce

a score (mean). As such, the score for each subscale can range from 0 - 4, on the

basis of the following calculations: Factor 1 = score / 5 items; Factor 2 = score / 4;

Factor 3 = score / 4; and Factor 4 = score / 3. Consequently, the complete

questionnaire can in theory return total scores ranging from 0 – 16. Higher scores

indicate that a participant is experiencing an increased intensity of emotions at a

specific time or in relation to a specific task. Therefore, the task could be considered

emotion-laden, suggesting that an individual is engaged or invested in the task.

Alternatively, lower scores indicate that a task is not emotionally engaging for an

individual, potentially due to a lack of contextual, situational, and affective


information or constraints (Headrick, et al., 2015; Kiverstein & Miller, 2015; Lewis,

2005).

Table 4.17 exemplifies the mean scores for each factor, and SLEQ total score

from the current sample of 342 respondents. It must be acknowledged that

participants also responded to 16 other items as part of the validation process, and

therefore these exemplar scores may not be entirely indicative of results for each

subscale. Nevertheless, statistical analysis of differences in the mean factor scores

revealed a significant main effect (F 2.05, 698.61 = 4.67, p < .05, ω =.78). Pairwise

comparisons returned significant differences (p < .05) between all factor

combinations except factors 2 and 4 (see Table 4.7 for mean and SD values). The

lack of significant difference in scores between factors 2 and 4 reflects the previously

discussed trend of negatively valenced items generally being selected less frequently.

Factor 2 and 4 scores also returned higher SD values than factors 1 and 3 suggesting

there was greater variability in participant ratings for the corresponding items.

Correlations between mean scores for each factor revealed statistically

significant and large positive interactions between factors 1-3 and 2-4, and a small

interaction for 2-3. Factors 1-4 returned a small negative interaction, and all other

combinations were insignificant (Table 4.7). Reflecting on the items loading to each

factor these correlational results appear to be quite logical. The large positive

correlations correspond with factors representing items of similar type and valence

that would generally be expected to fluctuate in unison (Shaver, et al., 1987).

Alternatively, the lone significant negative correlation indicates that scores for

factors 1 and 4 tend to mirror each other due to the contrasting emphasis on items

similar to ‘enjoyment’ and ‘annoyed’ respectively.


The final decision to make regarding the design of the proposed questionnaire

was the issue of labels to describe each factor or subscale. These labels would be

used to distinguish the four subscales and therefore needed to represent all loading

items effectively. It was considered important (where appropriate) to designate

labels that were distinctive to the SLEQ and the inherent focus on learning. As

mentioned previously the SEQ included the predetermined subscales of anxiety,

dejection, anger, excitement, and happiness (Jones, et al., 2005). Interestingly, none

of the items directly related to these labels (e.g. anxiety – anxious, dejection -

dejected) were the highest loading items in their respective subscales during the SEQ

development. Inspecting the factor loadings from Table 4.6 revealed that the highest

loading items for each SLEQ subscale were: 1 – ‘happy’; 2 – ‘nervous’; 3 –

‘satisfied’; 4 – ‘annoyed’. Despite these items loading highest on they do not

necessarily represent the overall theme for each group.

Taking into account the previous information it was decided to label the first

subscale as Enjoyment. As an item originating exclusively from the phase 1 list,

enjoyment had clear relevance to learning contexts. Enjoyment also was interpreted

to have stronger links with fun (also phase 1) and excited, when compared to the

highest loading item in happy (Jackson, 2000). Happy (or happiness) was also not

deemed relevant because it has been considered a lower intensity alternative to

another item, joy (Lazarus, 1991). Joy itself (and by association - happy) was found

to be unsuitable as it is a wide-ranging term, often referred to as a core ‘positive’

emotion, that somewhat related to several items from other subscales (Jackson, 2000;

Niedenthal, 2008). Both fun and excited were ruled out for the reason that they

simply did not represent the other items adequately.


The second subscale was termed Nervousness which reflected the predominant

factor loading of the item ‘nervous’ in both the initial and final structural model

designs. Regardless of the difference in factor loadings the remaining items were

either considered too broad (fear) or too specific (stressed, pressure) to adequately

represent the subscale (Öhman, 2008; Shaver, et al., 1987). The labels ‘anxiety’ or

‘anxiousness’ were also considered given that the item anxious was included in this

subscale in the initial design. However, this notion was rejected as a consequence of

anxiety relating to situations that are potentially threatening or dangerous and

creating insecurities in individuals (Lazarus, 1991; Raglin & Hanin, 2000), along

with the relatively weak observed factor loading.

The third subscale was labelled Fulfilment. A derivative of this term (fulfilled)

appeared in the happiness subscale in the preliminary SEQ, but was not retained as

an item in the final version of the SEQ (Jones, et al., 2005). Satisfied and successful

were deemed inappropriate due to links with the core positive concept of ‘joy’

(Shaver, et al., 1987). Accomplishment and achievement were both interpreted as

being unsuitable given the strong error covariance evidenced in model 2. Therefore,

designating either item as the subscale label was considered misleading due to their

established links.

The fourth and final subscale was designated the label of Anger to represent the

items ‘annoyed’, ‘angry’, and ‘frustrated’. These three items were also in the anger

subscale of the SEQ, however, frustrated was not retained in the final version (Jones,

et al., 2005). Some consideration was given for selecting frustration (frustrated) as

the subscale label given that it is a unique item to the SLEQ. However, on account

of factor loadings, and previous categorisations (Jones, et al., 2005; Lazarus, 1991;


Shaver, et al., 1987) it was concluded that anger provided a more appropriate

representation of the subscale as it included links to both annoyed and frustrated.

Drawing on the three previous phases of inquiry and analysis, the final version

of the Sport Learning and Emotions Questionnaire (SLEQ) is presented on page 113.

The four-factor design underpins two positive (Enjoyment - Fulfilment) and two

negatively (Nervousness – Anger) valenced subscales , which in turn represent 17

emotion items determined to be most relevant to learning contexts. Due to the focus

on emotions during learning contexts, and the manner in which items were identified

and refined, the SLEQ is suitable for implementation throughout learning sessions in

sport. The brevity and accessibility of this new emotions tool also affords the

opportunity to administer the SLEQ several times, and under the influence of various

constraints throughout a session. Therefore, the SLEQ is designed to facilitate

comparisons across various time scales alongside other measures (e.g. movement

analysis, performance outcomes, game plans, tactics) of value to athletes, coaches,

practitioners, and researchers.

Chapter 5: Implementing the SLEQ during learning in sport: Applied case examples 115

Chapter 5: Implementing the SLEQ during learning in sport:

Applied case examples

Overview

Following the design and development of the Sport Learning and Emotions

Questionnaire (SLEQ) in Chapter 4, the next stage of inquiry was to apply the tool in

applied learning contexts in sport. Therefore, Stage 3 of this thesis comprises of two

case examples demonstrating the use of the SLEQ in a team passing game (‘Endball’

– Example A), and a cricket batting task (Example B). These studies had two main

goals. Firstly, both of these examples were designed to demonstrate the

effectiveness of the SLEQ for measuring emotion intensity, in unison with action

(e.g. movement and performance), and cognition variables when key constraints

were manipulated. Despite previous investigations also including physiological

variables (e.g. heart rate, see Oudejans & Pijpers, 2009; Pijpers, et al., 2005; Pijpers,

et al., 2006), in the examples presented in this chapter a deliberate and theoretically

based decision not to include physiological variables was made with focus being put

on the crucial interaction between actions, emotions, and cognitions. This decision

was made on account of physiological responses often being associated exclusively

with select emotional terms such as anxiety, rather than more comprehensive

indications of emotion intensity incorporating both positive and negatively valenced

items (Jones, et al., 2005; Martens, Burton, et al., 1990). Furthermore, such

physiological measures are generally only markers of arousal, rather than direct

indications of emotion (Jones, 2003; Jones, et al., 2005). Therefore, the approach

adopted in these examples provided the opportunity to observe the dynamic between

overall emotion intensity, actions, and cognitions.

116 Chapter 5: Implementing the SLEQ during learning in sport: Applied case examples

The second goal of these examples was to highlight the relevance of the two

key principles of Affective Learning Design (ALD – see Chapter 3), specifically: (i)

the design of emotion-laden learning experiences that effectively simulate the

constraints and demands of performance environments in sport; (ii) recognising

individualised emotional and behavioural tendencies that are indicative of learning.

These principles highlight the importance of considering the influence of affective

constraints, and a holistic approach to emergent behaviour including emotion.

Example A involved the systematic manipulation of a task constraint

(possession time) during a session where the participants played the small-sided

invasion game ‘Endball’. Through the manipulation of this targeted constraint it was

expected that participants would show fluctuations in emotion intensity and actions

as they attempted to adapt to the changing task demands. The final game saw the

constraints return to the initial condition, in order to observe whether the preceding

manipulations produced any noticeable changes in emergent behaviour. Therefore,

the session design was deliberately constructed to facilitate learning over a short time

scale representative of experiences during a common physical education class, or

practice session. For the purposes of this example, measures of emotion (SLEQ) and

actions (game performance characteristics) were investigated to provide initial

insights regarding the effectiveness of the SLEQ. Hence, this example study was

considered as a critical ‘first step’ to implementing the SLEQ in applied sport

environments, along with demonstrating the second principle of ALD.

Example B provided the opportunity for an in-depth single case investigation

using the task vehicle of cricket batting. Space-time constraints were manipulated to

simulate increasing delivery (ball) speeds during a representative batter versus

bowler practice game scenario. This study design set out to increase the task

Chapter 5: Implementing the SLEQ during learning in sport: Applied case examples 117

demands from the perspective of the batter and observe related fluctuations in

emotion intensity (SLEQ), actions (kinematics and performance outcomes), along

with cognitions (via brief confrontational interviews during natural breaks in the

game). Therefore both principles of ALD were incorporated, in terms of including

affective constraints, and the recognition of individualised tendencies of actions,

emotions, and cognitions. The time scale of this example was again kept to one

session encompassing several constraint manipulations, simulating a typical practice

environment in sport.

Given the richness and volume of data available in both of these examples the

fluctuations between trials and/or conditions were readily able to be observed and

analysed. These fluctuations have often been neglected and overlooked in favour of

comparing mean values from a set or block of performances, often to the detriment

of understanding short term changes in key measures (Newell, et al., 2001). As

reported in the following sections, both examples provided evidence of linked

changes in actions and emotions (also cognitions in Example B) as constraints were

manipulated throughout various periods of the session. Longer term changes that are

indicative of learning were also observed across the sessions as a whole, reflecting

the notion of interacting time scales of learning (Granott & Parziale, 2002; Lewis,

2002; Newell, et al., 2001).

Despite the brief nature of the two sessions, both examples can be considered

representative of learning over short periods of time. This is further supported by the

functional adaptability displayed by individual participants and/or teams following

the manipulation to constraints discussed throughout the remainder of the chapter.

The methodological designs of both applied case examples satisfy the current aim of

demonstrating how the SLEQ can be implemented in applied sport learning contexts.

118 Chapter 5: Implementing the SLEQ during learning in sport: Applied case examples

Taking the following results and implications into account these studies form a

necessary step towards longitudinal investigations incorporating the SLEQ and

principles of ALD.

Chapter 5: Applied case example A 119

Chapter 5: Applied case example A

The influence of manipulated task constraints on the interaction

between emotions and actions in a small-sided invasion game

120 Chapter 5: Applied case example A

Abstract

Through the systematic manipulation of a targeted constraint this study aimed

to investigate the effectiveness of the newly developed Sport Learning and Emotions

Questionnaire (SLEQ) for tracking emotion intensity in an applied sport context.

The SLEQ was designed to reflect Affective Learning Design (ALD) principles, and

was implemented here to record the emotion intensities of individuals during a team

passing and possession game. Critically this investigation observed emotion

intensity alongside key game events (e.g. passes, touches, errors) under the influence

of manipulated constraints. Two teams of eight participants participated in four

games of ‘Endball’ (similar to Netball), with individual possession time restricted in

each case (3, 2, 1, and 3 seconds respectively). Manipulations in possession time

were found to be linked with emotion and action variable changes, particularly

during the transition from one game to the next (i.e. from the 2 second game – to –

the 1 second game). Significant correlational interactions were evident between

fluctuations in action variables (e.g. incomplete passes), and emotion subscale scores

(e.g. Anger) demonstrating the effectiveness of the SLEQ for measuring relevant

emotion intensities. These fluctuations were indicative of metastable regions created

by the systematic manipulation of constraints, where participants and teams were

forced to explore the environment for new functional states of emergent behaviour.

Therefore, this investigation demonstrates the value of the SLEQ for measuring

emotion intensity throughout an applied learning context, for understanding the

interaction between actions and emotions following the systematic manipulation of

constraints.


Introduction

The constraints based approach has received considerable attention recently as

an approach for understanding how a practitioner, coach, or teacher might

manipulate key features of a task to study or shape the behaviour of individuals in

sport (Chow et al., 2006; Davids, 2001; Renshaw, Chow, et al., 2010). This

approach is based on the three categories of constraints proposed by Newell (1986).

Organismic or Individual constraints represent characteristics of an individual in a

particular environment or situation. These characteristics include the physical size,

physiology, and strength of an individual and are inherently difficult to manipulate as

a consequence of such qualities being largely ‘internal’. Through previous sections,

emotions and cognitions have also been discussed as constraints reflecting an

individual’s interaction with and perception of environmental information (Headrick,

et al., 2015; Lewis, 2004; Pinder, et al., 2014). Environmental constraints can

include both physical (e.g. light, temperature, weather conditions) and social (culture,

peers, spectators) influences on learning or performance (Araújo, Davids, Bennett,

Button, & Chapman, 2004; Renshaw, Chow, et al., 2010). Environmental constraints

are often difficult to distinguish from the third category – task constraints (see,

Renshaw & Davids, 2014). Task constraints are generally more specific to a

particular situation and often relate to manipulations to rules, equipment,

instructions, and playing surface/area (Chow, et al., 2006; Hristovski, et al., 2006;

Newell & Ranganathan, 2010).

Task constraints are the most accessible form of constraints available to be

manipulated in sporting contexts. Examples include the manipulation of proximity-

to-goal and the presence of defending players in football interpersonal interactions

(Headrick, Davids, et al., 2012; Orth, Davids, Araújo, Renshaw, & Passos, 2014), the


impact on interceptive actions when implement characteristics are changed in tennis

and cricket (Buszard, Farrow, Reid, & Masters, 2014; Headrick, Renshaw, Pinder, &

Davids, 2012), and situation specific instructions in basketball (Cordovil, et al.,

2009). Studies such as these have provided valuable insights for how the physical

behaviours (including feelings, perception, decision making, and actions) of

individuals can be shaped through the systematic manipulation of key constraints.

The current project aims to build on previous constraints-led approaches by

considering how manipulations to key constraints might influence actions, in

conjunction with emotions, which have often been neglected (see Chapters 1-3).

The Sport Learning and Emotions Questionnaire (SLEQ), was specifically

designed for use during learning tasks in sport, and to address the need to better

understand the impact of manipulating specific learning tasks. Throughout the

previous chapters the theoretical underpinnings and structural design of the SLEQ

have been extensively discussed. This sport orientated tool is designed to be used at

several occasions throughout a session to document the intensity of select items

grouped to four distinct subscales (see chapter 4 – phase 2 and 3). The SLEQ

scoring system does not stipulate the type or valence of emotions experienced, rather

the overall intensity or ‘emotional engagement’ of an individual during a specified

time period.

The task constraint manipulated during this study was the maximum time a

participant could maintain possession of the ball during a common invasion game.

This constraint was chosen as it was a relatively simple variable to manipulate and

exemplified a manipulation that could readily be implemented by a coach or teacher

in a wide variety of sports and games (Moy, Renshaw, & Davids, 2014; Renshaw,

Chow, et al., 2010). Through systematic manipulation of this task constraint it was


anticipated that the performance of participants would initially be destabilised

leading to metastable periods (see, Chow, et al., 2011; Headrick, et al., 2015; Pinder,

et al., 2012). Therefore, participants would be forced to adapt to the change in

constraints and learn how best to satisfy their performance objectives within the new

task demands of the game. Observable changes in both actions (game events e.g.,

touches, errors, complete passes) and emotion intensity were expected to take place

during these metastable periods as participants actively searched for new or different

functional states (Hristovski, Davids, & Araújo, 2009; Pinder, et al., 2012). Further,

it was expected that SLEQ scores and performance variables would show evidence

of complimentary changes at key times (i.e. immediately following manipulation of

the constraint), supporting the previous theoretical arguments surrounding the

intertwined relationship between action and emotion. Taking into account the design

of the session (possession time systematically reduced before returning to the

original value – see methods) changes in emotion and game characteristics were

anticipated from the start to the end of game play. More specifically, it was

anticipated that the first and final games (identical possession times) would exhibit

different observable characteristics following the two manipulations in constraints

during the session. Essentially it was expected that changes in emotions and

performance would be shaped by previous experiences even though the task

constraints were the same for games 1 and 4. Potential improvements in

performance could be inferred through variables such as goals scored, errors, and

completed passes.


Methods

Participants

Eight male (n=5) and female (n=3) university students (mean age = 19.63 ± .74

years) voluntarily participated in the project as part of scheduled class activity

relating to an assessment project. Each prospective participant was briefed on the

nature of their involvement and asked to provide written informed consent prior to

commencing. The eight participants were provided with sufficient detail regarding

the study, but were not explicitly aware of the specific measures collected until data

collection commenced. Participants indicated current and/or previous experience in

the following sports: Football (soccer, n=3); Touch Football (n=2); Rugby League

(n=1); Boxing (n=1); and Athletics (n=1). Highest levels of representation in these

sports were: State (n=3); Regional (n=2); Club (n=3) and length of participation was

reported to be 6.63 ± 3.20 years. Therefore, none of the participants were

experienced in Netball or Basketball which share key characteristics with the Endball

game in this study. The participants were asked to self-select evenly matched teams.

This was deemed the best approach as the individuals were very familiar with each

other as they were in the latter years of their university course and were part of the

same cohort.

Procedure

Task design

Participants were required to play four short games of ‘Endball’ on an outdoor

grass surface. Endball is an invasion game where teams attempt to score goals by

having a team member take possession of the ball while he/she is standing within a

designated goal area at the respective ‘end’ of the playing area. This game was

chosen as it included many coordination modes for passing, catching, interpersonal

interactions, and important principles of play (e.g. time-space characteristics) also


exhibited in popular invasion game sports such as Netball, Basketball,

Football/Soccer and Hockey. Similarities in task constraints with Netball are

perhaps most prevalent, whereby: possession alternates from team-to-team freely

(i.e. following a goal or rule breach); a time limit is imposed for the duration each

individual possession by a player (3 seconds in Netball); and players cannot

move/run/dribble while in possession of the ball. As a result of these constraints

players/teams can only promote the ball towards the endzone through one or a series

of successful passes, each completed within the specified possession time limit. The

manipulated independent variable, in the form of a task constraint (or in ecological

dynamics terms a candidate control parameter2), throughout the four games in the

current investigation was the maximum time a participant could maintain possession

of the ball. Specifically, each individual possession time was initially constrained to

3 seconds, before systematically being reduced to 2 and then 1 second. For the final

game of the session, possession time returned to the original limit of 3 seconds. The

session design, task/possession time constraints for each game and response

collection occasions are presented in Figure 5.1 below.

Figure 5.1. Representation of session design with SLEQ and perception of difficulty (PoD) collection

occasions.

2 Informational variables that when manipulated have the potential to destabilise system organisation


Each of the four games took place on a playing area 20m in length and 12m in

width, with 2m deep ‘endzones’ used for scoring at each end. Two teams of 4

players participated and were identified by contrasting coloured bibs. Employing the

Game Intensity Index calculation (GII, see Chow, et al., 2013) the Endball game

equated to an index of 30 (GII = playing area (m2) / number of players), very similar

to that of Netball (GII = 33.2). Therefore, the intensity of the four games in terms of

players and space was considered appropriate for the purpose of the investigation.

Games consisted of two, three minute halves with two minutes rest between each

half, and five minutes between each game. A regulation (size 5) Netball was used in

all games, and a neutral referee enforced the relevant game rules with assistance in

terms of timing provided by members of the research team. Participants were asked

to play a warm up game for the purposes of familiarising themselves with the generic

game rules (no possession time restriction). This warm up also provided the

recorders (see later in this section) with a chance to familiarise themselves with the

key game characteristics that were to be observed.

Emotion and perception of difficulty

In order to observe how the manipulation of task constraints might influence

emotion throughout the session, the SLEQ was administered prior to the first game

(following warm up) and then following both half and full time for each game. On

each occasion participants were presented with a paper copy of the SLEQ and made

aware of the instructions and response scale. Participants were at no stage permitted

to review their responses from previous occasions, or communicate with other

players whilst completing the questionnaire. Following completion of the SLEQ

(30-60 seconds completion time), participants were also asked to indicate their

‘perception of difficulty’ in relation to playing in the respective half of each game.


This was recorded on a scale of 1 (very easy) to 10 (impossible) where difficulty

could relate to all aspects of the game (e.g. physical, skill, strategic demands).

Alongside the SLEQ, this simple measure was incorporated to provide individualised

and alternate perspectives on the influence of manipulations to the selected task

constraint.

Game analysis

The actions of each participant and team were recorded to establish if the

manipulated task constraint influenced specific game characteristics or events. Key

events were recorded for each team possession and included counts of: touches;

complete/incomplete passes; goals; dropped balls resulting in a turnover; time

violations; moving with the ball fouls; and out of bounds. Equivalent events were

recorded for individual participants within these team possessions, each by separate

pairs of recorders in the form of other students from the class. Pairs of recorders

independently observed and recorded the actions of their respective player/team

before coming together to cross check the events at the end of each half to ensure the

consistency and reliability of results. Inter-observer reliability returned mean

correlations of .88 ± .08 for the actions of the eight players, and .95 ± .02 for the

actions of the two teams. These reliability results met the required level, and in fact

were quite strong, given that events were recorded in real time (Van der Mars, 1989).

As mentioned previously, the recorders were given the opportunity to familiarise

themselves with key events during the warm up. In terms of the ‘result’, each game

was treated independently. That is, each game started with a score of 0-0.

Analysis

All results were collated by the research team and entered into spreadsheets for

initial analysis and inspection of possible patterns or trends in data. One-way


ANOVA tests were then performed to identify statistically significant differences

between teams throughout the session for goals scored, touches, passes (complete

and incomplete), combined errors, and perception of difficulty. Given that the

overarching aim of this study was to establish how emotion (SLEQ responses) might

link with actions during a learning session/lesson, the respective SLEQ and subscale

scores were the primary focus of the analysis. A series of one-way ANOVA tests

were performed to identify any differences between the two teams for overall SLEQ

and subscale scores at each measurement occasion. Effect sizes were also calculated

to accompany all ANOVAs using omega (ω), with the following classifications:

small = 0.10; medium = 0.30; large = 0.50) (Field, 2013). As a measure for

determining the change in SLEQ scores and game events between each half, z scores

were calculated. The z scores provided data specifying the change in SLEQ (and

subscales ENJ, NERV, FUL, and ANG) alongside the change in number of key

events (e.g. touches, goals), and perception of difficulty ratings during the respective

periods of play. Change in z scores were then subjected to a correlation (Pearson’s

correlation coefficient, r) analysis to determine any relationship between SLEQ score

and occurrence of game events following the previously mentioned manipulations to

task constraints (alpha < .05, 95% confidence intervals).

Results

Table 5.1 displays a summary of the observed game events, mean perception of

difficulty ratings (PoD), and mean SLEQ scores for both teams. Face value

inspection of these results shows that team number 2 were more dominant in terms of

game performance, outscoring team 1 in each of the eight halves of play, and also

accruing less errors (i.e. turnovers, time violations, moving with the ball fouls, out of

bounds). Mean SLEQ scores for each team are plotted alongside the frequency of


three game events (goals, touches, errors) in Figure 5.2. Following statistical

analyses, no significant differences were found between teams for mean goals

scored, or incomplete passes. Significant differences were found for the number of

touches in the first half of game 4 (F 1, 6 = 31.57, p < .05, ω =.89), in favour of team 1

(see Table 5.1 for mean ± SD values). In the same period team 1 also completed

significantly more passes (F 1, 6 = 6.40, p < .05, ω =.63). Team 1 was also found to

produce significantly more errors during the first half of game 1 (F 1, 6 = 10.71, p <

.05, ω =.74). Moving to PoD ratings, the only statistically significant difference

between teams was found during the second half of game 1, with team 1 players

reporting increased difficulty levels compared with team 2 players (F 1, 6 = 31.57, p <

.05, ω =.90). Responses to the SLEQ revealed significant differences in total SLEQ

score and Fulfilment (FUL) score during several periods of play, with the better

performing team (2) reporting higher scores on each occasion. Significant

differences in SLEQ scores were observed during: game 1 – 1st half (F 1, 6 = 12.04, p

< .05, ω =.76); game 2 – 2nd

half (F 1, 6 = 9.37, p < .05, ω =.72); game 3 – 1st half (F 1,

6 = 7.86, p < .05, ω =.68); game 3 – 2nd

half (F 1, 6 = 10.91, p < .05, ω =.74); and game

4 – 1st half (F 1, 6 = 8.05, p < .05, ω =.68). Fulfilment scores were significantly

different between teams during: game 1 – 1st half (F 1, 6 = 60.50, p < .05, ω =.94);

game 1 – 2nd

half (F 1, 6 = 45.18, p < .05, ω =.92); game 2 – 1st half (F 1, 6 = 10.24, p <

.05, ω =.73); and game 4 – 1st half (F 1, 6 = 15.78, p < .05, ω =.81).


Table 5.1. Summary of game events, mean perception of difficulty (PoD) ratings, and mean SLEQ

scores for each of the two teams.

PRE Game 1 Game 2 Game 3 Game 4

1st half 2nd half 1st half 2nd half 1st half 2nd half 1st half 2nd half

TEAM 1

Possessions - 9 9 11 8 9 7 9 9

Goals - 1 0 4 2 2 1 3 3

Touches - 28 34 59 42 46 26 44 38

Complete

Passes - 19 25 48 34 36 20 35 29

Incomplete

Passes - 0 0 7 3 6 2 3 1

Errors - 7 8 7 6 8 5 6 6

PoD - 5.25 ±

1.71

5.50 ±

0.58

5.50 ±

1.00

4.50 ±

1.29

4.75 ±

1.71

5.50 ±

2.38

4.75 ±

0.50

3.50 ±

0.58

SLEQ 6.16 ±

1.39

5.96 ±

1.54

5.67 ±

2.15

6.43

±1.41

7.16 ±

0.58

7.38 ±

0.44

6.81 ±

0.71

5.83 ±

1.14

7.20 ±

0.68

ENJ 3.40 ±

0.40

2.60 ±

0.40

2.35 ±

1.15

2.90 ±

0.81

3.00 ±

0.75

3.30 ±

0.70

3.10 ±

0.68

2.90 ±

0.77

3.35 ±

0.50

NERV 0.56 ±

0.43

0.88 ±

0.63

1.00 ±

0.54

0.63 ±

0.32

0.63 ±

0.43

0.81 ±

0.83

1.06 ±

0.83

0.31 ±

0.24

0.25 ±

0.20

FUL 1.95 ±

0.84

1.40 ±

0.54

1.15 ±

0.66

2.15 ±

0.68

2.70 ±

0.50

2.85 ±

0.19

2.15 ±

0.93

2.20 ±

0.16

3.35 ±

0.41

ANG 0.25 ±

0.32

1.08 ±

1.20

1.17 ±

1.29

0.75 ±

0.50

0.83 ±

0.58

0.42 ±

0.42

0.50 ±

0.43

0.42 ±

0.50

0.25 ±

0.32

TEAM 2

Possessions - 9 7 11 8 8 8 10 7

Goals - 5 4 5 4 5 3 6 4

Touches - 28 25 36 29 34 33 29 32

Complete

Passes - 19 18 25 21 26 25 19 25

Incomplete

Passes - 2 1 2 1 1 3 2 1

Errors - 4 3 5 4 3 5 4 3

PoD - 3.50 ±

1.0

3.25 ±

0.50

5.00 ±

0.0

5.25 ±

0.50

6.25 ±

1.89

6.75 ±

1.50

4.00 ±

1.63

3.75 ±

1.26

SLEQ 8.52 ±

2.10

9.15 ±

1.00

8.16 ±

0.31

8.25 ±

0.80

8.89 ±

0.97

8.90 ±

1.00

8.93 ±

1.06

7.73 ±

0.71

8.44 ±

1.29

ENJ 3.50 ±

0.20

3.40 ±

0.57

3.70 ±

0.26

3.20 ±

0.67

3.10 ±

0.66

2.75 ±

0.93

3.15 ±

0.70

3.35 ±

0.47

3.55 ±

0.34

NERV 1.44 ±

0.83

1.06 ±

0.77

0.56 ±

0.31

0.69 ±

0.24

0.69 ±

0.24

0.94 ±

0.83

0.88 ±

0.72

0.25 ±

0.35

0.13 ±

0.25

FUL 2.50 ±

0.53

3.60 ±

0.16

3.65 ±

0.34

3.45 ±

0.44

3.10 ±

0.66

2.80 ±

0.85

3.40 ±

0.43

3.30 ±

0.53

3.85 ±

0.10

ANG 1.08 ±

1.10

1.08 ±

0.74

0.25 ±

0.50

0.92 ±

1.00

2.00 ±

1.41

2.42 ±

1.66

1.50 ±

1.73

0.83 ±

1.11

0.92 ±

1.62


Figure 5.2. Mean team SLEQ scores plotted with: (a) goals scored; (b) touches; and (c) errors across

the session. *Significant differences in SLEQ scores between teams (p < .05)


Change in z score results are presented below in a collection of correlation

tables displaying the relationship between each observed variable (where

appropriate) throughout the 4 games. Table 5.2 and 5.3 present the correlation in

change of z scores for SLEQ and each of the subscales comparing Pre session

responses with those of game 1 – 1st half, and game 4 – 2

nd half (whole session)

respectively.

Table 5.2. Change in z score correlations between observed variables from Pre to Game 1 – 1st half

(SLEQ only).

ENJ NERV FUL ANG SLEQ

ENJ 1

NERV -.633 1

FUL .752*

[.204, .976] -.493 1

ANG -.701 .851*

[.657, .991] -.620 1

SLEQ .259 .395 .384 .315 1

* Significant correlation (p < .05)

Table 5.3. Change in z score correlations between observed variables from Pre to Game 4 – 2nd half

(SLEQ only).

ENJ NERV FUL ANG SLEQ

ENJ 1

NERV -.820*

[-.991,-.213] 1

FUL .826*

[-.420,.987] -.701 1

ANG -.484 .369 -.305 1

SLEQ -.039 .203 .101 .791*

[-.826,.991] 1


Tables 5.4 – 5.10 present correlations of change in SLEQ z scores along with

the range of game events recorded throughout each period of play. Table 5.11 reports

the same correlation relationships for the entire session.


Table 5.4. Change in z score correlations between observed variables from Game 1 – 1st half to Game 1 – 2nd half (3 second game).

ENJ NERV FUL ANG SLEQ PoD Touches Goals Comp

passes

Incomp

passes Errors

ENJ 1

NERV .074 1

FUL .651 -.402 1

ANG -.859*

[-.975,-.667] -.316 -.276 1

SLEQ .384 .821*

[.221,.986] .017 -.363 1

PoD .186 .666 -.515 -.541 .385 1

Touches -.324 .041 -.507 .203 .055 .047 1

Goals -.439 .556 -.710*

[-.970,-.093] .186 .143 .392 -.100 1

Comp

passes -.539 .625 -.618 .386 .284 .315 -.112

.897*

[-.276,1.000] 1

Incomp

passes -.048 -.373 .009 -.062 -.374 -.123 .629 -.588 -.612 1

Errors .647 .028 .624 -.517 .413 -.100 .136 -.785*

[-.996,-.403] -.689 .433 1



Table 5.5. Change in z score correlations between observed variables from Game 1 – 2nd half to Game 2 – 1st half (3 second game – 2 second game).

ENJ NERV FUL ANG SLEQ PoD Touches Goals

Comp

passes

Incomp

passes Errors

ENJ 1

NERV -.367 1

FUL .901*

[.690,.979] -.540 1

ANG -.817*

[-.985,-.313] .262

-.734*

[-.962,.098] 1

SLEQ .325 .540 .268 -.060 1

PoD -.572 .658 -.692 .542 -.007 1

Touches .041 -.204 .071 .502 .195 .166 1

Goals .734*

[.111,.979] -.302

.839*

[.278,.976]

-.823*

[-.987,-.410] .205 -.621 -.314 1

Comp

passes -.061 .221 -.032 .479 .666 -.098 .486 -.281 1

Incomp

passes .221 -.614 .463 .019 -.056 -.296 .410 .026 .191 1

Errors -.152 .417 -.391 .080 -.160 .742*

[.523,.992] .093 -.220 -.483 -.556 1



Table 5.6. Change in z score correlations between observed variables from Game 2 – 1st half to Game 2 – 2nd half (2 second game)


Comp

passes

Incomp

passes Errors

ENJ 1

NERV .055 1

FUL .728*

[.244,.943] .057 1

ANG -.729*

[-.968,-.030] .077 -.495 1

SLEQ .003 .345 .420 .483 1

PoD -.463 .052 -.654 .379 -.299 1

Touches -.481 -.051 -.101 .394 .224 .577 1

Goals .249 .600 .163 -.086 -.015 .441 .190 1

Comp

passes -.574 -.030 -.070 .417 .513 -.208 .341 -.580 1

Incomp

passes -.237 -.472 .011 .298 .096 .531

.817*

[.641,.974] .120 .074 1

Errors .207 -.368 .385 -.337 -.187 -.166 .320 -.035 -.200 .350 1



Table 5.7. Change in z score correlations between observed variables from Game 2 – 2nd half to Game 3 – 1st half (2 second game – 1 second game)


Comp

passes

Incomp

passes Errors

ENJ 1

NERV -.429 1

FUL .745*

[-.008,.980] -.616 1

ANG -.741*

[-.922,-.351] .177

-.787*

[-.976,.303] 1

SLEQ .242 .564 -.102 -.087 1

PoD -.235 .525 -.454 .089 .541 1

Touches -.236 -.443 .265 .215 -.572 -.635 1

Goals -.445 .198 .045 .202 -.125 -.401 .734*

[.432,.966] 1

Comp

passes .476 -.431 .026 -.030 -.235 -.197 -.149 -.653 1

Incomp

passes -.159 -.142 .329 .044 -.169 -.521

.811*

[.455,.967]

.886*

[.790,.999] -.540 1

Errors -.456 .130 .103 .068 -.250 -.373 .453 .756*

[-.194,.983]

-.707*

[-.920,-.088] .554 1





Comp

passes

Incomp

passes Errors

ENJ 1

NERV -.793*

[-.970,.335] 1

FUL .821*

[.429,.974]

-.750*

[-.982,-.014] 1

ANG -.826*

[-.995,-.477] .564

-.806*

[-.965,-.241] 1

SLEQ .736*

[.454,.980] -.414

.716*

[.186,.950] -.540 1

PoD -.181 -.037 -.318 .141 -.741*

[-.972,.034] 1

Touches -.162 .029 .216 .183 .322 -.677 1

Goals -.397 .460 .071 .171 .111 -.404 .446 1

Comp

passes .376 -.136 .096 -.129 .324 -.136 .163 -.516 1

Incomp

passes .306 -.258 .463 -.160 .654

-.764*

[-.998,-.066]

.850*

[.577,.977] .098 .533 1

Errors .024 -.223 -.021 -.021 -.289 .597 -.603 -.094 -.635 -.724*

[-.956,-.121] 1



Table 5.9. Change in z score correlations between observed variables from Game 3 – 2nd half to Game 4 – 1st half (1 second game – 3 second game)

ENJ NERV FUL ANG SLEQ PoD Touches Goals Comp

passes

Incomp

passes Errors

ENJ 1

NERV .060 1

FUL .432 -.380 1

ANG -.564 .671 -.482 1

SLEQ .440 .562 .511 .164 1

PoD -.648 .407 -.820*

[-.967,-.457] .504 -.359 1

Touches -.384 -.066 -.145 .056 -.241 .399 1

Goals .512 .572 -.074 -.066 .558 .135 .016 1

Comp

passes

-.730*

[-.992,-.081] -.073 -.065 .585 -.095 .179 .459 -.410 1

Incomp

passes -.371

.708*

[.034,.941] -.103

.847*

[.475,.975] .475 .319 .275 .076 .534 1

Errors -.086 -.447 -.090 -.515 -.579 .121 .343 -.364 -.297 -.458 1





Comp

passes

Incomp

passes Errors

ENJ 1

NERV .452 1

FUL .121 -.155 1

ANG -.217 .151 .245 1

SLEQ .435 .295 .738*

[-.106,.960] .295 1

PoD .103 -.072 -.739*

[-.972,-.477] -.315 -.641 1

Touches -.058 .113 -.559 -.281 -.548 .745*

[.345,.982] 1

Goals -.221 .131 -.679 -.452 -.709*

[-.942,-.043] .230 .389 1

Comp

passes .677 .179 .075 -.249 .125 .518 .373 -.391 1

Incomp

passes .144 .201 .526 .388 .369 -.210 .281 -.367 .319 1

Errors -.867*

[-.997,-.650] -.257 -.166 .363 -.157 -.202 -.039 .135

-.794*

[-.967,-.486] -.171 1



Table 5.11. Change in z score correlations between observed variables from Game 1 – 1st half to Game 4 – 2nd half (whole session)


Comp

passes

Incomp

passes Errors

ENJ 1

NERV -.278 1

FUL .718*

[-.055,.980] -.137 1

ANG -.484 -.191 -.353 1

SLEQ .166 .213 .579 .323 1

PoD -.793*

[-.995,-.032] .444

-.860*

[-.976,-.613] .501 -.155 1

Touches -.056 .310 -.292 -.528 -.588 .058 1

Goals .190 .461 .194 -.276 .170 .052 -.355 1

Comp

passes -.041 -.215 .158 -.301 -.199 -.324 .573

-.708*

[-.947,-.178] 1

Incomp

passes -.187 .096 .314 .163 .627 -.148 -.102 -.424 .397 1

Errors .302 .292 -.105 -.422 -.245 -.100 .454 .080 -.217 -.087 1



Discussion

Previous work in applied sport contexts has demonstrated how the

manipulation of key constraints can influence the actions and performance outcomes

of individuals. This current study aimed to investigate the influence of manipulated

constraints on actions, in unison with emotion intensity. Results revealed a range of

differences highlighting the effectiveness of manipulating constraints across the

session for both action and emotion variables.

Between team findings

Despite no statistical difference between teams in terms of goals scored, team 2

clearly won each game. Interestingly, in general, team 1 were found to complete

more passes and thereby more touches (significantly in game 4 – 1st half), while also

committing more errors (significantly in game 1 – 1st half). These findings highlight

the different style of play adopted by each team, with team 2 completing less passes,

but scoring more goals and recording fewer errors across most if not all periods of

play (see Table 5.1 and Figure 5.2). Perception of difficulty (PoD) results revealed a

significant difference following the 2nd

half of game 1 where team 1 again produced

higher mean ratings of difficulty. This observation reflects the event of team 1 being

outscored by 9 goals to 1 during this game.

SLEQ and Fulfilment subscale results between teams also revealed a number

of statistically significant differences across various periods of game play. Team 2

participants were found to report higher scores for both total SLEQ and Fulfilment

responses, suggesting that the dominance in goals scored was associated with

enhanced intensity of emotions relating to fulfilment or achievement, given that this

was the objective of the game. Therefore, it is important to note that despite no

evidence of significant differences in goals scored, the overall SLEQ and Fulfilment


subscale scores were able to distinguish between teams. Interestingly, team 2

reported higher (although not significantly) Pre session SLEQ and Fulfilment scores,

and maintained higher relative scores throughout all periods of play. This is with the

exception of a very slight mean deficit (Team 1= 2.85 ± 0.19: Team 2= 2.80 ± 0.85)

in Fulfilment score following the 1st half of game 3 (1 second game). The initial

difference in SLEQ score highlights individual differences in emotions experienced

by members of the two teams. Perhaps the difference may have been less

pronounced over the session if team 1 were more competitive in terms of goals

scored in one or several halves of play. These findings have some important

implications for practitioners such as teachers and coaches, and suggest that the

traditional ways of selecting teams based on perceived performance ability alone

may not be sufficient. As evidenced in this sample of participants, pre session

emotion intensities suggested that one team were more emotionally engaged than the

other. This information could be useful to a coach for the planning and design of

future sessions as a means of manipulating balance (or imbalance) between teams.

Overall correlation findings

Pre-Game – Game 1 (1st half) SLEQ

Overall (i.e. both teams) correlation results for changes in z scores revealed a

range of expected and unexpected findings. Contrasting changes in z scores from

pre-game observations with those following the first half of game 1 reveals

significant positive correlations between Fulfilment – Enjoyment, and Anger –

Nervousness (Table 5.2). These results were largely expected given that these

subscales broadly represent positive and negative items respectively, and would be

anticipated to fluctuate in unison (Diener, et al., 1985). Interpreting these findings in

relation to the session indicates that participants who experienced increased


fulfilment also experienced a higher intensity of enjoyment. Conversely participants

experiencing elevated intensities of anger also reported increased nervousness

intensities. These correlational results provide implications for practitioners in terms

of ensuring that individuals experience some success (i.e. Fulfilment subscale)

during a session, as it appears to be strongly related to enjoyment. To meet the

challenge of ensuring individuals experience some level of success, individualised

goals and objectives should be established providing the opportunity for competence

to be displayed across various tasks and skill levels within the same activity or

session (Chow, Davids, Button, & Renshaw, 2015; Renshaw, et al., 2012). Finally,

at this stage of the session, none of the SLEQ subscales were strongly correlated with

total SLEQ score.

Pre-Game – Game 4 (2nd half) SLEQ

Results presented in Table 5.3 compare correlations in the same variables

(SLEQ scores) across the whole session, from pre-game to the 2nd

half of game 4.

Once again a significant positive correlation between Enjoyment – Fulfilment is

evident. Taking into account emotion intensities spanning the session, a significant

negative correlation in z scores for Nervousness and Enjoyment is also present. This

relationship would generally be expected given that the individual items comprising

these two subscales on the whole represent contrasting experiences. For this set of

results another significant positive correlation was observed between total SLEQ

score and Anger, indicating that emotion items from this subscale (angry, frustrated,

annoyed) were strongly linked with the overall emotional intensity of the session.

Given the manipulation of constraints through this learning or training session was

intended to challenge and perturb the action and emotion states of participants, the

link between the Anger subscale and overall SLEQ score makes logical sense. These


findings again provide key practical implications for practitioners in terms of

recognising the increase in emotional intensity, and the primary subscale(s)

attributing to this increase in intensity across the session. Therefore, a teacher or

coach should take into account the possible changes in emotion intensities that may

emerge as a result of changes in constraints, and how such emotion intensities may

influence the actions of each individual.

Game 1 (1st half) – Game 1 (2

nd half)

Table 5.4 reports correlation results comparing changes in z scores from the

two halves of game 1, the original 3 second possession game. From this point on, the

results also include changes in z scores for game events along with PoD and SLEQ

results. For this comparison it can be seen that Anger – Enjoyment are again

negatively correlated, while Nervousness is positively correlated with overall SLEQ

score. This indicates that teams 1 and 2 might be experiencing higher intensities of

items from the Nervousness subscale due to the pressure or stress of feeling the need

to improve game performance (team 1) or maintain performance (team 2). At this

stage of the session changes in Fulfilment were found to be negatively correlated

with goals suggesting that participants in general did not solely base their perception

of fulfilment (success, achievement etc.) on scoring goals. This may have been due

to a mismatch in teams, resulting in a game that was not particularly engaging or

fulfilling (i.e. goals scored became inconsequential). These findings relate to the

challenge point framework where effective learning environments are considered to

include ‘optimal’ amounts of information to both challenge and engage individuals in

the learning process (Guadagnoli & Lee, 2004). In relation to goals, these results

also reveal that complete passes (positively) and errors (negatively) were

significantly correlated with changes in scoring. Overall these links seemed to have


face validity, however, as discussed earlier team 1 tended to complete more passes

and score fewer goals than team 2.

Game 1 (2nd

half) – Game 2 (1st half)

Correlations for changes to z scores from the second half of game 1 and the

first half of game 2 (2 second game) are presented in Table 5.5. Significant

correlations were observed for Fulfilment - Enjoyment (positive), Anger –

Enjoyment (negative), and Anger – Fulfilment (negative). Of note for this

comparison is the finding that the change in z score for goals was significantly

correlated with Enjoyment, Fulfilment (both positively), and Anger (negatively)

providing evidence to suggest that goals were strongly linked with the emotions

experienced by participants. As a possible consequence of the manipulation to task

constraints within this comparison (3 second – to – 2 second) errors and PoD were

also found to correlate positively at a statistically significant level.


nd half)

Comparisons between the two halves of game 2 (2 second game – Table 5.6)

once again reveal significant correlations between Fulfilment – Enjoyment (positive),

and Anger – Enjoyment (negative). Changes in z scores for incomplete passes and

touches were also found to correlate significantly and positively between the two

halves. This observation corresponds with both teams reducing the number of

touches and by association incomplete passes from the first to second half, implying

that the participants may have become more accustomed to the change in constraints

during the second half.


Game 2 (2nd


Similar correlations between Fulfilment – Enjoyment, and Anger – Enjoyment

were observed during the next comparison involving the second half of game 2 and

the first half of game 3 (1 second game). As with previous results during the change

from game 1 – to – game 2, a significant negative correlation between Anger –

Fulfilment is again reported (Table 5.7). This finding indicates that the change in the

temporal constraint resulted in increased intensities of anger and frustration along

with decreased fulfilment for some participants, in this case team 2. Team 1

represented the opposite trend with increased intensities of items on the Fulfilment

subscale and decreases for items on the Anger subscale. As anticipated, the change

to the 1 second game brought about several other linked changes in game events.

Changes in z scores for goals scored correlated positively with touches, incomplete

passes, and errors suggesting that participants were forced to change their style of

play in order to maintain or increase their goal scoring frequency. Positive

relationships with incomplete passes and errors in particular indicate that participants

were willing to make more mistakes for each goal scored. This is further supported

by the significant positive correlation between incomplete passes and touches.

Finally, the change in errors and complete passes were negatively correlated which is

no surprise given the change in constraints and the requirement to complete passes in

order to score a goal.


nd half)

Changes in z scores following each of the halves during game 3 revealed both

Enjoyment and Fulfilment subscales to be significantly correlated to the three other

subscales and total SLEQ score (Table 5.8). As observed on previous occasions

Enjoyment correlated with Anger and Nervousness negatively, and Fulfilment


positively. The same trends were observed for Fulfilment between the respective

subscales. Total SLEQ score was positively correlated with both Enjoyment and

Fulfilment subscales providing evidence to suggest that the constraint of 1 second

possession time influenced overall emotional engagement by means of how intensely

participants experienced items relating to enjoyment and fulfilment. Perception of

difficulty ratings were observed to negatively correlate with both total SLEQ score

and incomplete passes again reflecting the influence of the 1 second task constraint.

Overall changes in the frequency of incomplete passes were also significantly

correlated with touches (positively) and errors (negatively).

Game 3 (2nd


Transitioning from the 1 second game (game 3 – 2nd

half) back to the 3 second

game (game 4) corresponded with a negative correlation between PoD and

Fulfilment (Table 5.9). This can be interpreted as reflecting the perceived decrease

in difficulty as a result of the longer time permitted for possession of the ball. As

discussed earlier, the first half of game 4 saw significant differences between teams

in terms of more touches and completed passes for team 1, while team 2 recorded

higher scores for Fulfilment and total SLEQ. These results suggest that returning to

the original constraints of the 3 second game produced a range of team and

individual specific changes, including an immediate increase (though not statistically

significant) in goals scored for both teams. During this period of the session several

significant interactions were also observed between SLEQ subscales and game

events. Enjoyment negatively correlated with complete passes suggesting that an

increased number of passes does not necessarily relate to increased intensities of

items such as happy and excited. Changes in z scores for incomplete passes were

also found to correlate positively with both Nervousness and Anger. This finding


suggests that incomplete passes were potentially considered more influential than

errors (drops, rule violations) in terms of elevated intensities of items such as

frustrated, pressure, annoyed and stressed.


nd half)

Interpreting results between the two halves of game 4 reveals that errors were

negatively correlated with Enjoyment and complete passes which was to be expected

(Table 5.10). Fulfilment was found to be positively linked with total SLEQ score

and negatively correlated with PoD. These findings suggest that returning to the 3

second constraint in game 4 led to many participants experiencing increased

intensities of items relating to Fulfilment as a consequence of decreases in perceived

difficulty. Changes in the number of touches were also linked to PoD positively

indicating that participants may have considered an increased number of passes to

elevate difficulty perceptions. Finally, goals and SLEQ were found to be negatively

correlated, again indicating that an increased number of goals scored did not

necessarily reflect the emotional engagement of participants, whether positively or

negatively valenced as discussed previously.


nd half)

The final set of correlation results presented in Table 5.11 concerns the change

in observed variables from the first half of game 1 through to the second half of

game 4. Therefore, these results reflect change in z scores across the session

encompassing the manipulation of the identified task constraint. For this comparison

Fulfilment – Enjoyment were again positively correlated, while PoD was negatively

correlated with both of these SLEQ subscales. These findings appear quite intuitive

and indicate that the manipulation of constraints during the session may have led to

participants reporting higher intensities of Enjoyment and Fulfilment in unison.


Contrasting decreases to PoD ratings also suggest that the constraint of 3 seconds

possession time was less challenging or demanding for participants after

experiencing both 2 and 1 second manipulations. As noted during previous

comparisons the change in complete passes was found to negatively correlate with

goals scored. This implies that an increased number of complete passes was not

imperative for goal scoring, hinting at the notion of quality passes over pure quantity.

Implications

From a theoretical perspective these findings support the intertwined

relationship between emotions, and action as modelled and conceptualised earlier

(Headrick, et al., 2015; Lewis, 2004; Lewis & Todd, 2005). Following systematic

manipulations of constraints across the four games, changes in emotion and action

measures were observed, indicating that the participants were forced into periods of

metastability (Hristovski, et al., 2009; Kelso, 2012; Pinder, et al., 2012). During

these periods (e.g. the transition from the 3-to-2 second game) the players could be

considered to have self-organised their actions and emotions in an attempt to be

functional in relation to the new task demands (Hristovski, Davids, Passos, &

Araújo, 2012).

This preliminary investigation of emotion and action relationships in sport has

exemplified the effectiveness of the SLEQ as a tool for recording emotion intensity

at various time points. The design of the SLEQ provided the opportunity to gauge

the response of players following each half of play, all of which had the potential to

correspond with changes in emotion. In tandem with notational analysis of key game

events the SLEQ provided rich information that a coach, teacher, or sport practitioner

could use in evaluating the effectiveness of skill development or game play


interventions on account of emotional engagement and physical performance

characteristics.

Limitations and future research

An acknowledgeable limitation of this investigation was the small sample of

participants and the data collection period being limited to one class. However,

when reflecting on the findings of this investigation of eight participants, the richness

of the data provides justification for further research to focus on in-depth analyses of

small groups rather than striving for ‘statistical power’ in the form of mass cohorts of

participants. This approach would be particularly beneficial for intervention studies

with elite sporting populations where by definition, large samples sizes are not

attainable (Pinder, Headrick, & Oudejans, 2015). A single lesson could be

considered representative of short time scales of learning, however, in order to

strengthen the findings a longer time scale project would be required, tracking the

key emotion and action variables through several consecutive classes or sessions

typical of school physical education programmes. The method of collecting game

event data also had its limitations (particularly for individual participant results)

given that all recording was completed ‘live’ as part of the class project. Subsequent

investigations would benefit from the use of video footage documenting each

constraints manipulation for later analysis. This would allow a more in-depth

analysis of performance, for example, by affording analysis of passing strategies or

patterns changed within and between games. The imbalance in Pre session SLEQ

scores in favour of team 2 also provides some scope for future work in terms of

attempting to pick or select teams that are closely matched in emotion intensity,

rather than on perceived performance capabilities alone. Furthermore, due to the aim

of evaluating the usefulness of the SLEQ in this study, the findings focussed on the


SLEQ and subscale scores along with game performances, and largely neglected the

role of cognitions. Future studies should incorporate cognitions to examine the

intertwined relationship between all three concepts to adopt a more holistic approach

(see Applied case example B). Finally, despite the short time length of the games,

the influence of fatigue must be recognised as a potentially confounding factor on

game events and SLEQ scores.

Conclusion

In summary this preliminary study has demonstrated the practicality and

effectiveness of the SLEQ for tracking emotion intensity over a select period of time.

Interpreting SLEQ results with key game events provided initial, yet strong support

for the principles of ALD in practice while also supporting the theoretical approach

developed throughout the thesis.

Chapter 5: Applied case example B 153

Chapter 5: Applied case example B

Action, emotion, and cognition interactions during a cricket batting

practice task

154 Chapter 5: Applied case example B

Abstract

This study adopts a single case design to investigate actions, emotions, and

cognitions using the task vehicle of cricket batting. To effectively incorporate and

observe the interaction between these variables an Affective Learning Design (ALD)

approach was implemented. During a simulated game scenario, ball delivery speed

was manipulated by systematically reducing the distance from which the bowler

released the ball. Throughout the five manipulations in speed, a battery of measures

were collected to observe interactions between actions (e.g. movement

characteristics, performance outcomes), emotions (intensity using the SLEQ), and

cognitions (confrontational interviews to establish intentions and goals). Findings

revealed a range of interlinked relationships between variables, particularly when

transitioning between delivery speeds (metastable regions). In particular, foot

movement distances, Enjoyment subscale scores, perception of achievement ratings,

and runs scored measures revealed interlinked tendencies. Therefore, this in-depth

individualised investigation provides valuable implications for practitioners

regarding the design and study of emergent behaviour during learning events in

sport.


Introduction

The concept of Affective Learning Design (ALD), discussed extensively in the

previous chapters, advocates for: (i) the design of emotion-laden learning

experiences that effectively simulate the constraints of performance environments in

sport; and (ii) recognising individualised emotional and coordination tendencies that

are associated with different periods of learning (Headrick, et al., 2015; Pinder, et al.,

2014; Renshaw, et al., 2014). The investigation discussed during this section

incorporates both of these principles using a cricket batting task as vehicle to observe

how the manipulation of task constraints influences actions, emotions, and cognitions

throughout a targeted skill development task which was representative of a game

situation.

In terms of creating emotion-laden learning experiences, previous work has

largely focussed on the advantages to learning outcomes when training under

pressure and the constraints of induced performance anxiety in a range of tasks

(Oudejans, 2008; Oudejans & Nieuwenhuys, 2009; Oudejans & Pijpers, 2009, 2010).

In several related investigations by these authors, participants training under the task

constraint of induced anxiety were found to more successfully maintain their

performance levels in high anxiety performance conditions, compared with those

who trained in low anxiety conditions. These tasks included throwing darts when

positioned at different heights from the ground (Oudejans & Pijpers, 2009, 2010),

handgun shooting at cardboard targets versus live opponents with marking cartridges

(Oudejans, 2008), and Basketball shooting with/without being watched and evaluated

during practice sessions (Oudejans & Pijpers, 2009). Such findings have clear

implications for how affective task constraints can be manipulated during learning or

skill development tasks, reinforcing the importance of considering how behaviour


and performance outcomes can be constrained by emotional and cognitive

information.

A limitation of this previous body of work (and others) is that induced anxiety

is the clear focus rather than affect or emotion overall. In order to monitor

individualised emotion and cognition tendencies across various periods of learning,

in unison with actions or performance outcomes, a more holistic approach is

required. The previous chapters have detailed the targeted development and utility of

the Sport Learning and Emotions Questionnaire (SLEQ) for tracking the emotional

intensity or engagement of individuals over several time points. This tool provides

an overall score for emotional intensity along with four distinct subscale scores for

Enjoyment, Nervousness, Fulfilment, and Anger. The SLEQ has also been

developed within sport contexts, for use within sport contexts, as opposed to many

currently used tools that have been adapted from clinical based instruments. A

suitable method for identifying cognitions alongside emotions is the use of

confrontational interviewing techniques. This approach is based on an individual

reviewing their own performance (often with guidance) to continuously identify key

cognitive aspects relating to the situation (Gernigon, et al., 2004; von Cranach &

Kalbermatten, 1982). Therefore, the use of targeted questioning can establish the

thoughts, intentions, goals, and tactics of individuals to track how manipulations in

constraints might alter responses over time (McPherson, 2008).

The task vehicle of batting in cricket has been studied previously from a

variety of perspectives relating to the manipulation of constraints and information

available to performers (for reviews see, Portus & Farrow, 2011; Stretch, Bartlett, &

Davids, 2000). Examples include the manipulation of visual information through

occlusion techniques (Müller & Abernethy, 2006), comparing the use of ball


projection machines with live bowlers on batting actions (Pinder, Davids, et al.,

2011a; Pinder, Renshaw, & Davids, 2009), and attunement to and the influence of

using bats with different weight characteristics (Headrick, Renshaw, et al., 2012), to

mention a few. Typically these studies have been conducted in situations not

particularly representative of the demands and characteristics of match conditions at

times due to the constraints imposed by data collection equipment / techniques and

the need to ‘control’ the environment. This is reflective of many cricket batting

sessions which are commonly carried out in practice nets with limited simulation or

representation of situation specific information present in match conditions, such as

score, field placements, and interaction with direct opponents (Pinder, Davids, et al.,

2011b; Pinder, Renshaw, et al., 2011; Renshaw, Chow, et al., 2010).

This current investigation adopted a task design that sampled and selected

many informational variables that would normally be present in a match situation.

Key to the session design was the implementation of a game scenario generating

dynamic contextual information that aimed to emotionally engage both the bowler

and batter in an interpersonal contest, while also providing the opportunity to observe

and collect various variables pertaining to actions, emotions, and cognitions. These

characteristics reflect key principles and recommendations of ALD in applied sport

settings in terms of incorporating and recognising the influence of affective

constraints (Headrick, et al., 2015). Related conceptualisations in cricket have led to

the development of coaching tools such as the ‘battle zone’, effectively a small-sided

training game specific to cricket which represents key aspects of a game within a

confined playing area (Renshaw, Chappell, et al., 2010; Vickery, Dascombe,

Duffield, Kellet, & Portus, 2013).


The distinguishing feature of this current study is the systematic manipulation

of simulated delivery speeds to examine any intertwined changes in actions

(movements and performance outcomes), emotions (SLEQ scores), and cognitions

(confrontational interview responses, and rating scales). Following previous findings

it was expected that increased delivery speeds would result in adapted movement

patterns, such as shorter front foot step lengths (Pinder, Davids, et al., 2011a;

Renshaw, et al., 2007; Stretch, Buys, Du Toit, & Viljoen, 1998). Emotion scores

were expected to increase in intensity as the session progressed reflecting the

implementation of more challenging task demands. Cognitions were hypothesised to

be related to ongoing perceptions of success or failure during the session as the

situation of the game-like task changed. It was also envisaged that key variables

representing each of actions, emotions, and cognitions would show interrelated

tendencies at pivotal times during the session. This characteristic would indicate

periods of metastability (see Chapters 2 and 3) where actions, emotions, and

cognitions of the participant had been perturbed by the manipulation of task

constraints.


Methods

Participant

The participant for this exploratory study was a 19 year old, male, cricket

batter. In terms of ability, the participant was a top order batter who had represented

Australia at U-19 level and was a junior-professional with an Australian State. Prior

to data collection commencing the participant was provided with informed consent

information, and written consent was obtained.

Procedure

Task design

The cricket batting task for this investigation required the participant (batter) to

face 10 overs (6 deliveries each) from a known bowler of slightly lower playing

standard (Club premier / 1st grade standard) with space-time incrementally

manipulated to simulate faster deliveries (distance between batter and bowler

manipulated). The bowler was selected on account of being familiar to the batter,

which provided the opportunity to observe how manipulating space-time demands of

deliveries might simulate a faster bowler, while retaining similar bowling action

characteristics. This is typical of match and practice situations where a batter will be

familiar with the bowler, the characteristics of their deliveries, and tactics. A known

bowler of slightly lower standard to the batter was also preferred to address safety

concerns given the manipulation in simulated delivery speeds. To further maximise

safety the batter was required to wear full personal protective equipment at all times,

and the bowler was specifically instructed not to bowl deliveries bouncing above

chest height. However, the batter was not made aware of this instruction to ensure

that the types of deliveries were not easily predictable.

Data collection was carried out at an indoor cricket net practice facility familiar

to both the batter and bowler (synthetic grass). The design of the facility allowed the


task to take place in a relatively open space (50m x 20m) when compared with

traditional cricket nets. Both the bowler and batsman were instructed to perform in

the context of the opening 10 overs of a one-day match (50 overs a side), assuming

the roles of an opening bowler and batsman respectively (also their current team

roles). Therefore, the batter aimed to score as quickly as possible without being

dismissed, and the bowler aimed to restrict runs and ultimately get the batter out. To

simulate match conditions an umpire was in position and the bowler had the

opportunity to place fielders (in the form of small nets) in positions permitted during

the opening 10 overs of a game (Renshaw & Davids, 2004). A regulation four-piece,

white leather cricket ball was used throughout the session.

Task manipulation

In order to manipulate the space-time characteristics of deliveries, the length of

the pitch was modified following every second over (5 sets of 2 overs). This

involved systematically moving the popping crease line markings, stumps, and

umpire closer to the batter according to predetermined distances equating to 5 km/h

increments in delivery speed (Table 5.11). To determine the range of adjustments

the normal delivery speed of the bowler was initially assessed with a sports radar gun

(Stalker Radar, Texas) positioned on a tripod between the umpire and stumps. In this

case the delivery speed of the bowler across 6 practice deliveries was found to be

approximately 115 km/h (Mean = 115.5 ± .33 km/h). Therefore, the initial condition

was set at ~115km/h off the full length pitch distance (popping crease – to – popping

crease = 17.68 m). This base speed and distance was subsequently used to calculate

the remaining task manipulations presented in Table 5.11, by matching the time

available.


Table 5.11. Delivery speed calculations

Using the bowler’s standard speed of 115 km/h the pitch length was

manipulated by matching distance and time relationships to create the following

conditions:

a) Set 1 (overs 1-2) Full pitch = simulated speed ~115km/h

b) Set 2 (overs 3-4) Pitch shortened by 0.75m = simulated speed ~120km/h

c) Set 3 (overs 5-6) Pitch shortened by 1.25m = simulated speed ~125km/h

d) Set 4 (overs 7-8) Pitch shortened by 2m = simulated speed ~130km/h

e) Set 5 (over 9-10) Bowler asked to increase speed to ~ 120km/h - Pitch shortened

by 2m = simulated speed ~135km/h

Delivery speed

km/h 115 120 125 130 135

m/s 31.94 33.33 34.72 36.11 37.50

Full Pitch (0m) Distance 17.68

a 17.68 17.68 17.68 17.68

Time 0.55a 0.53 0.51 0.49 0.47

0.25m Distance 17.43 17.43 17.43 17.43 17.43

Time 0.55 0.52 0.50 0.48 0.46

0.5m Distance 17.18 17.18 17.18 17.18 17.18

Time 0.54 0.52 0.49 0.48 0.46

0.75m Distance 16.93

b 16.93 16.93 16.93 16.93

Time 0.53b 0.51 0.49 0.47 0.45

1m Distance 16.68 16.68 16.68 16.68 16.68

Time 0.52 0.50 0.48 0.46 0.44

1.25m Distance 16.43

c 16.43 16.43 16.43 16.43

Time 0.51c 0.49 0.47 0.45 0.44

1.5m Distance 16.18 16.18 16.18 16.18 16.18

Time 0.51 0.49 0.47 0.45 0.43

1.75m Distance 15.93 15.93 15.93 15.93 15.93

Time 0.50 0.48 0.46 0.44 0.42

2m Distance 15.68

d 15.68

e 15.68 15.68 15.68

Time 0.49d 0.47

e 0.45 0.43 0.42


The decision to cease shortening the pitch at 2m was made for safety reasons

and the potential that ball bounce characteristics may become unrepresentative at

closer distances. For each condition, the position of the popping crease was re-

marked using tape and the stumps maintained at the regulation distance of 1.22m

behind. The popping crease is a line marked at both ends of the pitch and some part

of the bowler’s front foot must land behind this line to constitute a legal (fair)

delivery. The umpire and radar were also kept consistent and repositioned

proportional to the change in pitch distance. Indicated speed from the radar gun was

recorded for each delivery to determine how consistently the bowler was meeting the

designated speed. Before each set (2 overs of 6 balls) the batter was made aware of

any changes to pitch length, however, he was not explicitly made aware of the

delivery speed being simulated.

Actions and performance outcomes

To capture the batter’s physical actions in terms of movement characteristics

and cricket specific outcomes a range of measures were collected. Two-dimensional

kinematic movement data was recorded in order to obtain front-foot movement

distances (toward or away from the bowler), and maximum bat lift heights following

previous work (Headrick, Renshaw, et al., 2012; Pinder, et al., 2012). To capture

these events a high speed camera (Casio EXILIM Pro EX-F1: 300 FPS) was

positioned 11m perpendicular to the plane of action in line with the popping crease at

the batter’s end following previous procedures and recommendations (Bartlett, 2007;

Headrick, Renshaw, et al., 2012; Pinder, Davids, et al., 2011a). Key events were

analysed using Kinovea 2D motion analysis software (version 8.15). A second

camera (Basler Ace: 300FPS) was positioned to capture the batter front on, looking

over the top of the bowler and umpire. This camera was elevated 4.33m above the


ground and 47.2m from the popping crease at the batter’s end. Footage from this

camera was used to review ball bounce (line/length), shot type and ball direction in

unison with the side-on footage.

A validated measure of bat-ball contact, Quality of contact (QoC, Müller &

Abernethy, 2008), was completed post session by three members of the research

team (Australian level 1-3 accredited coaches) to determine whether contact was

considered: 2 = Good; 1 = Poor; or 0 = no contact. A subjective measure of

Forcefulness of Bat-swing (FoBS, Mann, Abernethy, & Farrow, 2010) was also

completed to give an indication of the intent involved in each shot (Pinder, et al.,

2012). This measure (originally: 2 = high; 1 = moderate; and 0 = low forcefulness)

was somewhat modified to better account for different types of bat swing

characteristics and was scored on a scale of 0 - 4: 0 = leave/no bat swing; 1 =

defensive shot; 2 = ‘checked’ shot; 3 = incomplete swing; 4 = full swing. Both of

these subjective measures provided some description of the shots performed, but

were not considered primary variables for analysis. Inter-observer reliability was

performed on all 60 trials for both QoC and FoBS. Correlations were strong between

all three observers for both QoC (Observer 1-2, r = .82; 1-3, r = .81; 2-3, r = .88),

and FoBS (Observer 1-2, r = .88; 1-3, r = .90; 2-3, r = .96). Runs (0s, 1s, 2s, 3s or

4s) were recorded live by the research team with input from the batter and bowler.

Where disagreements between the batter and bowler could not be resolved, the

research team made the ultimate decision. This occurred on only two occasions

during the session. Observations of shot type and direction were recorded live during

the session and later confirmed via reviews of the footage.


Emotions

In order to determine any emotional changes resulting from the manipulation in

delivery speeds the Sport Learning and Emotions Questionnaire (SLEQ) was

employed. The SLEQ was completed prior to the session and following each of the

ten overs bowled. The Positive and Negative Affective Schedule (Watson, et al.,

1988) was also completed prior to and following the session as a comparative

measure of emotions. Both of these tools were implemented by the same member of

the research team.

Cognitions

As a simple measure of how fast the batter considered deliveries in an over a

‘Perception of Speed scale’ was presented. This scale ranged from 0 – 10 and

included labels commonly associated with cricket delivery speeds to accompany the

batter’s rating (see Appendix F). Following each of the 10 overs, the batter was also

asked a set of targeted questions inquiring into his thoughts regarding the previous

and upcoming overs where relevant (Gernigon, et al., 2004; McPherson &

MacMahon, 2008). Further questioning about the session overall took place

following the completion of the 10 overs. The audio of these questions and answers

was recorded and transcribed for later reference (see Appendix G). Following initial

inspection of these responses a ‘Perception of Achievement’ scale was devised based

on responses to the first question “Who do you think won that over, yourself or the

bowler?” Responses to this question were found to lend themselves to 5 point scale

as follows: 1 – Him (i.e. bowler); 2 – Probably him; 3 – Even; 4 – Probably me; 5 –

Me (i.e. batter).

Analysis

Actual delivery speed (regardless of pitch length manipulation) was monitored

throughout the session to determine how consistent the bowler was in meeting the


designated speeds. All emotion and action results were collated by the research team

and entered into spreadsheets for initial analysis and inspection of possible patterns

or trends in data. One-way ANOVA tests were then performed to identify

statistically significant differences for action variables across the 10 overs. The same

method was used to compare values for these variables across the six balls of each

over (i.e. ball 1 versus ball 5). Brown-Forsythe statistics were reported if

homogeneity of variance was violated. Effect sizes were also calculated to

accompany all ANOVAs using omega (ω), with the following classifications: small

= 0.10; medium = 0.30; large = 0.50) (Field, 2013). Key action variables are as

follows:

- 1st foot movement: Two-dimensional distance of the first front foot

movement towards or away from the direction of the bowler relative to the

batter’s stance/ready position. This movement was found to be an

individualised tendency of the batter.

- 2nd

foot movement: Distance from the position of the 1st foot movement to

the position when the shot was completed.

- Total foot movement: Absolute movement distance of the front foot (1st

and 2nd

combined) during the shot.

- Bat lift height: Maximum height reached by the toe end of the bat during

the back lift/swing measured from the ground.

- Runs: Runs scored per over or ball number as judged by the research team

in consultation with both batter and bowler.

Perception of speed, emotion (SLEQ and subscales), foot movements, bat lift

height, runs scored, and perception of achievement for each over were all


transformed to z scores as a method for determining any linked changes between

variables across the session. Change in z scores were then subjected to a correlation

(Pearson’s correlation coefficient, r) analysis to determine any relationship between

SLEQ scores and movement/performance variables following the manipulations to

task constraints (alpha < .05, 95% confidence intervals).

Cognition results from the brief confrontational questions were transcribed and

tabulated for each respective over. Given the exploratory nature of this single case

study, the question and answer responses were not subjected to full qualitative

analysis. Key aspects of these short responses were, however, included with other

results (including QoC, FoBS, shot type, and direction) to create an over-by-over

summary of how the innings progressed to provide an overview of how emotions,

cognitions and actions interacted.

Results

During the first four sets (8 overs) the bowler was found to be very consistent

in meeting the target speed of ~ 115km/h: Set 1 (115.08 ± 1.08 km/h); Set 2 (114.5 ±

1.24 km/h); Set 3 (113.33 ± 2.64 km/h); Set 4 (114.58 ± 1.88 km/h). However, the

target of ~ 120km/h in Set 5 proved to be a challenge (112.92 ± 1.88 km/h) with

deliveries ranging from 110 – 116km/h. Despite the target speed not being met this

set remained part of the study given the richness of action, emotion and cognition

findings reported in following sections in relation to the simulated contest between

batter and bowler.

Over comparisons

Comparisons of action and movement results between the 10 overs revealed

statistically different findings for both 2nd

foot movement (F 9, 35 = 2.71, p < .05, ω

=.42) and total foot movement (F 9, 21.68 = 2.55, p < .05, ω =.40). Post Hoc


Bonferroni comparisons revealed overs 7 and 9 to be significantly different for both

variables: 2nd

foot movement (over 7 = 7.71 ± 6.03 cm; over 9 = 66.54 ± 35.05 cm, p

< .05), total foot movement (over 7 = 20.78 ± 12.60 cm; over 9 = 75.04 ± 47.03 cm,

p < .05). Figure 5.4 presents total foot movement for each over alongside SLEQ, and

total runs scored. The three remaining variables: 1st foot movement (F 9, 16.73 = .625,

p > .05, ω =.22); bat lift height (F 9, 24.79 = .581, p > .05, ω =.24); and runs (F 9, 32.16 =

1.93, p > .05, ω =.32) revealed no significant differences.

Ball number comparisons

Comparison for balls 1 – 6 revealed no significant differences in any of the

specified variables: 1st foot movement (F 5, 26.09 = 1.70, p > .05, ω =.28); 2

nd foot

movement (F 5, 36.66 = 1.35, p > .05, ω =.20); total foot movement (F 5, 54 = .45, p >

.05, ω =.27); bat lift height (F 5, 12.69 = .35, p > .05, ω =.30); and runs (F 5, 54 = .67, p >

.05, ω =.21). Figure 5.6 displays the ball-by-ball results for runs scored and SLEQ

scores prior to the session and following each over to visually illustrate run scoring

patterns throughout the session.

Change in z score correlations

Change in z score correlation results for the session are presented in Table

5.12. Variables compared are the SLEQ and subscale scores (Enjoyment,

Nervousness, Fulfilment, and Anger), perception of speed (PoS), perception of

achievement (PoA), foot movement distances, bat lift height and runs scored per

over. From these results it can again be observed that total foot movement and to a

lesser extent 2nd

foot movement distances correlated strongly with other variables

from both action and emotion categories. Figure 5.5 visually represents trends

between SLEQ subscale scores throughout the course of the session. PANAS

scores/scales for positive and negative affect are also presented highlighting the


increase in both scales across the duration of the session. Positive (26 – 31 =

16.13%) and negative (18 – 20 = 10.0%) affect increased between these two time

points compared with total SLEQ (4.3 – 9.42 = 54.34%), Enjoyment (1.4 – 3 =

53.33%), Nervousness (1.5 – 2.75 = 45.45%), Fulfilment (1.4 – 3 = 53.33%), and

Anger (0 – 0.67 = 100.0%).

Question and answer responses

Responses to the brief confrontational interviews following each over and the

session overall are presented in full in Appendix G (p. 295), with selected key

responses included in the set and over summary figures that follow.

Figure 5.4. Mean (combined 1st and 2

nd) foot movement, total runs, and SLEQ score throughout the

10 over session. * significant difference in foot movement (p <.05)


Figure 5.5. SLEQ subscale scores during the session and PANAS scores Pre and post (over 10)

session.


Figure 5.6. Ball-by-ball runs scored and SLEQ score at the end of each over


Table 5.12. Change in z score correlations for the session. * Significant correlation (p < .05)

ENJ NERV FUL ANG SLEQ PoS PoA

1st foot

move

2nd

foot

move

Total foot

move

Bat lift

height Runs

ENJ 1

NERV -.026 1

FUL .336 -.509 1

ANG -.265 .541 -.696*

[-.917,-.054] 1

SLEQ .207 .635 -.248 .789*

[.386,.956] 1

PoS -.307 .231 -.471 .334 .074 1

PoA .365 -.341 .630 -.534 -.220 .158 1

1st foot

move -.290 .175 .085 .385 .441 -.549 -.534 1

2nd

foot

move

-.732*

[-.934,-.430] .056 -.644 .538 .003 .305 -.657 .273 1

Tot foot

move

-.718*

[-.942,-.459] .109 -.527 .578 .137 .025

-.722*

[-.945,-.193] .588

.928*

[.762,.984] 1

Bat lift

height -.130 -.069 -.064 -.106 -.217 -.216 -.616 .176 .276 .207 1

Runs .648 -.112 .634 -.522 -.037 -.019 .875*

[.453,.979] -.424

-.739*

[-.931,-.300]

-.755*

[-.969,-.292] -.523 1


Over by over summary

The following pages (Figure 5.7 – 5.11) present a timeline of the session to

illustrate how the contest between batter and bowler emerged over the course of the

10 manipulated overs. Each figure presents ball-by-ball and over summaries of the

respective sets/pairs of overs. The specific type of cricket shots (e.g. BF – back foot;

BFD – back foot defence; FFD – front foot defence) are listed alongside runs scored,

QoC, and FoBS to provide a detailed description of each action. Visual

representations (wagon wheel) of where each delivery was hit in relation to the

batter’s position, and how many runs were scored are also presented. Deliveries that

were not played at, or missed by the batter were not included on the wagon wheels

(Note: batter was left-handed – off-side = left of page; on / leg-side = right of page).

Solid lines represent shots played with the batter’s weight primarily on the front foot

(moving towards the bowler - FF), and dashed lines represent back foot shots (away

from the bowler - BF) (Stretch, et al., 2000; Woolmer, Noakes, & Moffett, 2008). In

terms of the batter’s cognitions, the over by over game plan was extracted from the

interview transcripts to provide an indication of the strategy and intentions involved

throughout the session.


Figure 5.7. Set 1 (over 1 – 2) summary


Discussion

The aim of this chapter was to establish any potential links between actions,

emotions, and cognitions when space-time constraints were manipulated in a game-

based cricket batting task. Reviewing the results from the simulated practice game

suggests that manipulation in delivery speed revealed observable dynamic changes in

the intertwined relationship between SLEQ scores, cognitions (game plans, tactics),

movement characteristics, and performance outcomes of the participant. Therefore,

the holistic and representative nature of this situation specific learning / skills session

provides evidence to support the concepts of Affective Learning Design (ALD), and

Representative Learning Design (RLD) advocated throughout the previous and

following chapters (Headrick, et al., 2015; Pinder, Davids, et al., 2011b; Renshaw, et

al., 2014).

Action findings

From the 10 overs faced the batter was found to score 67 runs without being

dismissed, which represents a more than satisfactory start to a one-day match innings

(run rate of 6.7). Movement and performance outcome variables observed

throughout the session provided several indications regarding the emergent

behaviour of the batter following manipulations to task demands. Distances (forward

or back) covered by the batter’s front foot were found to vary significantly between

overs 7 and 9 for both the 2nd

foot movement and combined (1st and 2

nd) movements.

Over 7 was the first over of the simulated ~130km/h condition, which ultimately was

the fastest speed replicated given the speed requirements during the final two over set

were not met. Interestingly during over 8, the second over in the ~130km/h set, foot

movements increased to an average distance comparable with overs from previous

slower speed sets. Therefore, it can be concluded that during the initial exposure to


this speed the batter adapted his actions to shorter front foot movements, likely due

to the decreased time available to act in a functional manner (complete a successful

shot) (Hristovski, Davids, Araújo, & Passos, 2011). Shorter foot movement

distances could also be interpreted as the batter ‘freezing’ his degrees of freedom to

maintain functionality (score runs) under the influence of perceived increases in task

demands (Bernstein, 1967; Berthouze & Lungarella, 2004; Newell, Broderick,

Deutsch, & Slifkin, 2003). From a cricket batting perspective, reducing the degrees

of freedom signifies more compact, less extravagant movements, for the intention of

scoring runs in a select few areas of the field, or simply surviving (i.e. not being

dismissed). Furthermore the batter appeared to embrace system degeneracy, by

continuing to exhibit functional performances despite observable changes in

movement organisation (Edelman & Gally, 2001; Komar, Chow, Chollet, & Seifert,

2015; Pinder, et al., 2012).

The observation of foot movement distance increasing again during the second

over at this speed (130km/h - over 8) indicates that the batter had adapted and

attuned to the demands of the ‘new’ task, even within the space of 6 deliveries

(Headrick, Renshaw, et al., 2012). Hence, it could be suggested that some of the

previously ‘frozen’ degrees of freedom were released (Hong & Newell, 2006). This

finding is supported by previous cricket batting research findings where step length

changed as a result of manipulated constraints (Headrick, Renshaw, et al., 2012;

Pinder, Davids, et al., 2011a; Pinder, et al., 2009). The absence of significant

findings for first foot movement distance and bat lift height suggest that these

movement characteristics were less susceptible to the influence of increased delivery

speed. Both these characteristics could also be considered as consistent preparatory

movements of the batter throughout the session (Woolmer, et al., 2008).


Emotion findings

Scores for total SLEQ and each of the subscales indicated that emotional

intensity and/or engagement fluctuated during the session with a general trend of

increasing scores. The trends of each subscale plotted in Figure 5.5 show that the

positively orientated subscales of Enjoyment and Fulfilment tended to increase

during the first two sets (~115km/h and ~120km/h) before plateauing for much of the

remaining overs. Nervousness and Anger, the negatively orientated subscales,

appeared to show a decreasing trend during the first 3 – 4 overs respectively before

increasing to a peak at the end of over 9 (first over of intended ~135km/h condition).

These reductions could be interpreted as the batter becoming more attuned to his

environment (Headrick, Renshaw, et al., 2012; Hristovski, et al., 2006; Oudejans,

Michaels, & Bakker, 1997). Over 10 scores showed Nervousness and Anger to

decrease (or at least stabilise) and both Enjoyment and Fulfilment increase,

potentially as a result of the designated speed simulation not being reached. This

implies that the batter (who was under the impression that speed would be

increasing) might have perceived that the speed had not in fact increased from the

previous set and adopted a previously stable (attractor) state of organisation

accordingly (see Figure 5.4).

Comparisons with PANAS, a pre-established affect measurement scale,

revealed clear similarities with SLEQ scores when considering Pre and Post session

responses. PANAS scales for both positive and negative affect increased by 16%

and 10% respectively, while SLEQ and individual subscale scores increased by at

least 45% over the same time period. Returning to Figure 5.5 these trends can be

observed, with the obvious limitation of only comparing Pre and Post measures is

highlighted in terms of the fluctuation in SLEQ scores throughout the session. This


point reinforces a key advantage of the SLEQ given that it was designed specifically

to be completed multiple times during throughout a session in relation to a specified

task or activity. Furthermore, as stated in Chapter 4 and by Jones et al. (2005), the

PANAS was originally intended for use in daily living contexts and only considers

the broad categories of positive and negative affect, rather than the four distinct

subscales derived specifically for sport of the SLEQ. The SLEQ also differs in that it

provides indication score of overall emotion intensity that indicates an individual’s

engagement with a task as opposed to separate scores corresponding with positive or

negative affect. Therefore the SLEQ score reflects the importance of overall

emotional engagement in a task rather than focussing on scores for individual

emotions or contrasts between positive and negative orientations (Jarvilehto, 2000a,

2009; Lewis, 2004).

Cognition findings

The perception of speed (PoS) and perception of achievement (PoA) scales,

derived to provide indications of how the batter rated the delivery speed (0-10) and

winner of each over (1-5), also fluctuated following delivery speed manipulations

(Figure 5.7 – 5.11). The batter’s PoS ranged from a score of 6 (medium – fast) in the

first set (~115km/h) through to 9 (express) in the final set of overs, which were

intended to simulate deliveries of approximately 135km/h, but ultimately were closer

to 130km/h as in set 4. Interestingly, the batter rated set 4 (overs 7 & 8) slightly

slower at 8 – 8.5 (fast) than overs 9 & 10 in the final set, even though the delivery

speed did not increase. Part of the reason for this difference may have been the

safety requirement of informing the batter that the pitch was shortened and the

bowler asked to bowl faster. Alternatively, the batter may have attuned to key

bowling action or body language information which was perceived as pre-emptive of


changes in delivery speed characteristics (Müller & Abernethy, 2012). Of further

interest is the increased intensity of Fulfilment scores between overs 9 and 10. This

suggests that even though the batter thought the deliveries were faster in this set, he

attuned to key perceptual information and self-organised to exploit the constraints

functionally (i.e. scored more runs in over 10 than over 9).

Perception of achievement ratings revealed that the batter believed he had

competed successfully against the bowler for all overs apart from overs 1 and 9.

During these overs 1 and 2 runs were scored respectively. However, in very

different periods of the simulated game scenario. Over number 1 was the beginning

of the game and therefore it would be expected (traditionally) that a batter shows

some restraint as they begin to build an innings and become attuned to the bowler’s

actions and pitch characteristics. Over 9 was the first over of the final set, where the

batter was anticipating an increase in speed from the previous set while also

attempting to maintain the run-rate from a 50 over game perspective (see Figure 5.6).

The final set also signalled the end of enforced field restrictions from a game context

perspective. Referring to Figures 5.4 and 5.5 it can be seen that Anger and

Nervousness subscales, along with overall SLEQ score peaked following over 9,

while Fulfilment and Enjoyment decreased from the previous overs. The Anger

scale (angry, frustrated, and annoyed) showed a substantial increase in intensity at

this point, suggesting the batter had not met his own performance goals or

expectations, thus reflecting the low PoA rating.

Combined findings

Bringing the action, emotion and cognition observation findings together

provided an insight as to how these individual measures intertwined throughout the

session. Correlations of changes in z scores are presented in Table 5.12 and show


several strong (and significant) relationships. Moving from left-to-right, the first key

relationship is that of Enjoyment with both 2nd

foot movement, and total foot

movement distances. These strong negative correlations indicate that shorter foot

movements were linked strongly with higher intensity scores on the Enjoyment

subscale. Therefore in this case shorter foot movements are not indicative of poor

performance, instead suggesting an efficiency of movement for the intention of

‘accessing’ identified scoring zones through a shift in body weight (Woolmer, et al.,

2008). In fact the highest Enjoyment scores were reported for overs 4, 5, 7, and 8

which also corresponded with the shortest average values for total foot movements

(Figure 5.4). A strong negative relationship was also found between change in

Fulfilment and Anger scores, which seems quite logical given the opposing trends in

these scores particularly between overs 8 and 9 as discussed previously. Building on

this point is the strong positive correlation between the Anger subscale and total

SLEQ score for the session. This relationship implies that an increase in SLEQ score

was often accompanied by an increase in scores for the items in the Anger subscale

(angry, frustrated, and annoyed). Together these relationships indicate how emotion

intensities and actions can interact, which provides further support for principles

advocated by ALD (Headrick, et al., 2015).

Additional correlations were found when contrasting changes in the cognition

derived variable, perception of achievement, alongside runs scored, and total foot

movement. A strong positive relationship was observed between PoA and runs

scored which seems logical given the batter’s overarching aim for the session was to

score runs. The negative relationship between total foot movement and PoA

indicates that when the batter believed he had won an over his total foot movement

distance tended to be shorter. This notion is further supported by the next set of


positive relationships between runs scored and both 2nd

and total foot movements.

Interpretation of these interactions suggests that overs where more runs were scored

also exhibited shorter foot movements. Finally, as would be expected from earlier

findings the positive interaction between 2nd

and total foot movement variables was

also very strong given that the 2nd

foot movement contributed substantially to total

movement.

Therefore an indirect finding from these correlation results is as follows;

Increased runs linked with increased PoA increased PoA corresponded with

decreased total foot movement decreased total foot movement corresponded with

increased enjoyment (and vice versa). This synchronisation between variables

representing action (runs, total foot movement), emotions (Enjoyment), and

cognitions (PoA) further supports the integration of ALD principles (specifically

principle ii) into learning environments in sport. Understanding how observable

variables such as these interact during learning events provides rich information for

coaches, teachers, and practitioners regarding the influence of manipulated

constraints on the emergent behaviour of an individual (Pinder, et al., 2015).

Consequently, these findings highlight the need for coaches to include situational

information (affective constraints) that specifies a particular context in a match or

competition setting, and attempt to monitor the emergent behaviour from a holistic

approach, not restricted to outcomes or physical observations (Headrick, et al., 2015;

Renshaw, et al., 2009).

Over by over summaries

The cricket specific information presented in Figures 5.7 – 5.11 provides a

detailed description of how the simulated game progressed throughout the session.

Of particular interest is how the range and type of shots played develops over the


course of the session. The initial 2 – 3 overs revealed a restricted range of shots

(mostly Back foot defence) concentrated between the positions of point and mid-off,

and scoring few runs ('building an innings', see Woolmer, et al., 2008). As the

session continued the range of shots expanded with the exception of over 6 and 9

where the batter also indicated the over was equal, or won by the bowler

respectively. The expansion of shots and run scoring culminated with over 10 where

each of the 6 balls were scored off including three boundaries (4s) to different areas

of the field.

When revisiting previous results involving the significantly shorter total foot

movement in over 7 compared with 9, it can be deduced that this tendency was

functional for the aim of the batter for a number of reasons. Firstly, over 7 produced

8 runs (two boundaries) indicating that the batter did “hit the bad balls” as indicated

in the game plan reported after over number 6. Following over 7, the bowler chose

to make three changes to the field positions which would suggest his strategy was no

longer deemed functional for his goals. This equates to the batter perturbing the state

or balance of the game and forcing the bowler to explore other options (metastable

period) (Hristovski, et al., 2009; Kelso, 1990, 2012). These ideas are further

supported by the PoA score (5) for the over, indicating that the batter believed he

won the over convincingly. The five shots where bat-ball contact was made during

this over were also made on the front foot which implies that the batter was shifting

his weight towards the bowler, but with relatively short (efficient and functional for

the task) foot movements. These findings somewhat contradict previous cricket

batting literature where longer front foot movement distances are considered to be

preferred or “technically correct” (Woolmer, et al., 2008). Finally, all four SLEQ


subscales, and therefore total SLEQ score, increased following this over indicating

that the emotional engagement of the batter was further enhanced.

Reviewing the same observations for over 9 (highest average front foot

movement) shows that only 2 runs were scored by the batter, and the game plan

revolving around “try to attack him a bit more” was not executed well. This may

have been a result of the 3 field changes made prior to over 8 perturbing the game

plan of the batter and restoring some balance or stability between the two players as

with interpersonal dynamics in sport research (e.g., Headrick, Davids, et al., 2012;

Passos, et al., 2008). The PoA score (1) confirms that the batter believed the bowler

had convincingly won the over. Thus, the findings from this over strongly support

the consideration of SLEQ intensities alongside cognitive game plans to interpret

how emotions link with the goals and objectives of a task. Five of the six shots in

over number 9 were played from the back foot including the two scoring shots,

which travelled to the same general area of the field. Overall SLEQ score for this

over was the highest for the session with Anger and Nervousness subscales peaking,

and Enjoyment and Fulfilment dropping off compared with previous overs. These

findings indicate high overall emotional intensity was largely due to increased Anger

(angry, frustrated, annoyed) and Nervousness (nervous, stressed, fear, pressure)

scores. Finally the game plan in relation to over 10 reflects the batter’s thoughts

about the failure of his strategy with the previous over and his revised plan that he

“Might go a bit more reserved. Stay a bit more still”. This change in intentions or

strategy appeared to be successful given that over 10 produced a session best of 15

runs.


Implications

Building on the theoretical implications from the Endball study (Example A)

this more detailed individualised study provides further evidence of the critical

relationship between action and emotion, along with cognition. By chronicling

measures relating to each of these concepts in a representative cricket batting task the

manipulation of space-time information was found to create periods indicative of

system metastability (Hristovski, et al., 2009; Kelso, 2012). Incorporating principles

of ALD into the task also provided the opportunity to observe the interpersonal battle

between the batter and bowler due to the competition or game context simulated

(Pinder, Davids, et al., 2011b). The sampling of information and match

characteristics into this practice task was found to be very effective given that many

of the batter’s cognitions were very specific to the state of the game. On several

occasions these cognitions (e.g. game plans from overs 7-9) were pre-emptive of

fluctuations in emotion intensity, movement characteristics and performance

outcomes reflecting self-organising tendencies of these distinct yet clearly

interrelated measures.

Practical implications arising from this investigation again focus on the

applicability of ALD principles to systematically manipulated sport environments

(Headrick, et al., 2015). The inclusion of more detailed analysis techniques provided

evidence of individual movement characteristics (i.e. foot movements) fluctuating

alongside outcome based measures (runs, shot type) providing future scope for

studying emergent behaviour at various scales of analysis. The SLEQ was again

found to be highly effective in assessing the emotional engagement of the batter and

compared well with a previously used scale in PANAS, particularly given that the

SLEQ was suitable for tracking rich emotion responses throughout the simulated


game. In terms of implications for cricket batting practice the design of this

preliminary investigation appeared to be successful in implementing shorter pitch

distances to simulate faster delivery speeds, while also maintaining an environment

with sufficient contextual and situational information to engage both batter and

bowler.


A potential limitation that must be acknowledged involves the manipulation of

the pitch length to simulate different delivery speeds. As a result of decreasing the

pitch length it can be assumed that the bowler was forced to release the ball later in

the bowling action in order to bounce the ball at approximately the same length as

when bowling from a full pitch distance. Therefore, based on previous work

investigating the perception of information during interceptive actions (e.g. Müller &

Abernethy, 2006; Pinder, Davids, et al., 2011a; Renshaw & Fairweather, 2000;

Stone, Maynard, North, Panchuk, & Davids, 2015), the batter may have perceived

the balls delivered from the reduced pitch lengths as short balls (i.e. balls normally

played off the back foot on a full length pitch). As such the movement behaviour

and shot choices displayed by the batter might have also been influenced by the

delayed ball release point. However, despite the potentially confounding influence

of this change in bowling action characteristics, the proportion of back foot shots

played did not increase as pitch distance was reduced (see Figures 5.7 – 5.11). This

suggests that the batter still anticipated, and crucially, was able to perceive and

discriminate between either short or full deliveries as the pitch distance was

shortened. Furthermore, the finding that front foot movement distance was on

average longest during over 9 (i.e. fastest target delivery speed and equal shortest

pitch distance) indicates that the batter was still attempting to play both front and


back foot shots, rather than ‘camping’ on the back foot. Therefore, the manipulation

of pitch length may have constrained critical bowling action characteristics that a

batter attunes to. However, the role of simulated delivery speed and interlinked

affective constraints should not be discounted as influences on movement behaviour

and batting performance outcomes.

Reflecting on this study further there are some key considerations that could be

incorporated in subsequent investigations. Firstly, in this case the bowler did not

achieve the desired maximum delivery speed and therefore the last set of overs was

somewhat compromised even though some key findings were evident during this

period of the session. Next, if the study was to be repeated the actions, emotions,

and cognitions of the bowler should also be collected to investigate perceptions,

intentions, and emergent behaviours alongside those of the batter. Such measures

would also be of great interest in relation to the above discussion about changes in

bowling action characteristics respective to the length of the pitch. The requirement

of the bowler to consistently meet designated speed and length (not bounce above

chest height) requirements also has the potential to eliminate tactical deliveries such

as slower/off pace balls, cutters, and bouncers. Incorporating these refinements,

future work should look to study a range of batter – bowler combinations to examine

the effectiveness of this approach for different skill levels, ages, and playing styles

(e.g. spin bowlers, lower order batters). Finally the possibility of completing this

task outdoors on a ‘real’ grass cricket pitch and field should be explored to further

enhance the simulation of a match scenario.

Conclusion

Overall using cricket as a task vehicle to explore the interaction between

actions, emotions, and cognitions was successful in that several links between these


categories of measures were observed as the batter was exposed to increased space-

time demands following the manipulation of pitch length. Theoretical and practical

implications indicate the effectiveness of an ALD approach for incorporating

contextual information and affective constraints into learning tasks, and considering

actions, emotions, and cognitions in unison when critiquing emergent behaviour in

relation to performance goals. Furthermore, manipulation in space-time demands

was found to be an effective method for studying changes in emotions and linked

emergent behaviour characteristics in a cricket batting task.

Chapter 6: A conceptual model of affective learning design 191

Chapter 6: A conceptual model of affective learning design

The previous chapters have discussed the theoretical underpinnings of

Affective Learning Design and provided two examples of how ALD principles can

be adopted in sport settings. By incorporating these ideas and findings the following

chapter proposes a conceptual model of ALD that intends to provide coaches,

teachers, and sport practitioners with a theory driven framework for applying these

principles.

192 Chapter 6: A conceptual model of affective learning design

Abstract

Successful learning and practice environments should effectively represent and

sample information that is present in the simulated performance environment;

therefore, acknowledging the relationship between an individual and the environment

is imperative. The influence of affect in learning design is often neglected or

suppressed because it is perceived as a source of unwanted variability. Here it is

argued that a principled integration of emotion in learning experiences should be

embraced and considered as an integral component of designing representative

learning tasks. By adopting an ecological dynamics approach a holistic model of

‘Affective Learning Design’ that embraces the role of affect (emotion) in the design

and implementation of learning tasks is proposed. To support the conceptual model,

two applied examples are discussed providing practical insights into each of the key

model phases. Therefore, this model provides implications for the practical

application of affective learning design principles in practice and future research,

recognising the interaction between affect, behaviour, and cognition across different

time scales of analysis.


Introduction

Learning is an inherently emotional experience where an individual must

frequently deal with success, failure and associated physical and psychological

challenges (Davids, 2012; Pinder, et al., 2014; Seifert, Button, et al., 2013). For

example, the experiences of learning to walk, read or drive a car are everyday

activities that pose unique challenges for individuals at different stages of life

(Dolan, 2002). In the case of sport, recognising the role of emotion during periods of

learning is essential to understand the manner in which individual athletes perceive

and act in specific environments. Australian Test cricketer Ed Cowan (2011, p. 11)

described the process of learning to hit boundary shots (for T20 games) as an

experience that involved dealing with the emotions of initial inadequacy and a fear of

failure even though he was an experienced professional cricketer. Each individual

arrives at a learning experience with specific capabilities that must be modified or

adapted in order to meet the demands of the new task (Kelso, 2003; Schöner, et al.,

1992). Therefore, designing learning experiences must adopt an individualised

approach and consider concepts including perception, intentions, attention, actions,

cognitions, and emotions (Davids, 2012; Kelso, 2003).

Why it is important to consider emotion in learning

Historically, much like movement variability, emotion during learning (and

performance) has often been considered as unwanted noise (Davids, et al., 2003;

Seifert & Davids, 2012; Smith & Thelen, 2003). Therefore, sport scientists, teachers,

and coaches often attempt to remove or suppress emotion from learning and

experimental environments to facilitate controlled and predictable behaviour (de

Weerth & van Gert, 2000). Suppressing emotions during learning suggests a focus

on manufacturing predictable performance outcomes rather than allowing individuals


to explore the environment and learn from their own positive and negative

experiences. By distancing emotions from learning, the role of individual differences

in learning is also diminished, preventing individuals from establishing personalised

behavioural characteristics in particular environments and situations (Davids, et al.,

2003; Seifert & Davids, 2012). Furthermore, the role of emotion during learning has

often been neglected because emotion-laden responses are considered irrational or

instinctive, and therefore perceived as negative (Hutto, 2012; Lepper, 1994).

Emotionless responses made from a purely rational perspective have been described

as ‘cold cognition’, whereas emotion-laden responses are viewed as ‘hot cognition’

(Abelson, 1963; Lepper, 1994). The expression of ‘sit on your hands’ in relation to

choosing a move in a game of chess is an example of the view that it is necessary to

suppress or remove emotions in order to make rational decisions (i.e. cold cognition)

(Charness, et al., 2004). Jarvilehto (2000a, p. 53) summarised this idea when writing

that “emotion was regarded only as a factor bringing chaotic or irrational elements to

the computational process”. Crucially in sporting contexts, learners are often not

afforded this ‘thinking’ time and must act impulsively based on the interaction

between their perceptions of the task and pre-existing physical, cognitive, and

emotional capabilities (Davids, 2012).

Progress in understanding emotions has been limited by traditional cognitive

thinking perpetuating the debate over the pre-eminence of cognition over emotion;

where cognitions of events are thought to result in emotional reactions (Headrick, et

al., 2015; Lewis & Granic, 2000; Renshaw, et al., 2014). These ideas highlight that

this outdated reductionist approach to understanding emotions by cognitivists has

hampered the ability to model relations between goals, emotions and emotion

regulation (Lewis & Granic, 2000). However, some psychologists have recently


begun to acknowledge the advantages of nonlinear dynamic systems in explaining

behaviour and this has led to the emergence of a dynamical systems perspective of

emotional development (Lewis, 1996; Lewis & Granic, 2000). Yet to be seen in the

literature, however, is a principled exploration of the role of emotions during

learning for sports performance.

Tasks that are emotion-laden are considered to facilitate a ‘deeper’ engagement

for learning and performance (Jones, 2003; Solomon, 2008). Indeed, emotional

engagement is seen as being essential for effective learning (Pessoa, 2011). Emotions

should also be considered as part of learning experiences on account of the

relationship between emotion, behaviour and cognition, which form a trichotomy of

feeling, acting, and knowing respectively (Breckler, 1984; Hilgard, 1980). Behaviour

not only encompasses overt actions which in their simplest form involve moving

towards or away from an object, but also verbalisations about acting and intentions to

act. Cognition has been conceptualised as the knowing of the object including

thoughts, beliefs and perceptual reactions (Ajzen, 1989; Breckler, 1984; Hilgard,

1980).

While cognition and emotions are often treated as separate entities (Mathews &

MacLeod, 1994; Pessoa, 2008; Schwarz, 2000), the three components of affect,

behaviour and cognition have been shown to influence each other to form emergent

appraisals of experiences (Frijda, 1993; Lewis, 1996). During the appraisal of an

experience the interaction between cognition and emotion in particular is critical

(Frijda, 1993). From this approach, cognition and emotion are considered to

influence each other in a coupled feedback loop where cognitions bring about

emotions, and emotions shape cognitions (Lewis, 2004). This cyclical interaction

underpins the self-organisation of cognitive, emotional and perceptual aspects of


experiences to form what Lewis describes as emotional interpretations (Lewis, 1996,

2000a). Emotional interpretations (EI) are stable attractor states that form when

emotional and cognitive characteristics become linked or synchronised, and

subsequently facilitate the emergence of an action tendency (behaviour) (Lewis,

2000a). For example, the observable differences in movement behaviour

(completion time, exploratory movements) when induced anxiety was manipulated

through wall climbing traverses of different heights (Pijpers, et al., 2005), or critical

differences in movement patterns between expert and beginner ice climbers due to

variations in individual constraints (i.e. intentions, perceptions - Seifert, Button, et

al., 2013).

Evidence suggests that emotions are ‘triggered’ subconsciously before

perceptions or cognitions, meaning that an appraisal of an experience cannot be

completed without the influence of emotion (Jarvilehto, 2001; LeDoux, 2000; Lewis,

2000a). Emotional experiences are also considered to be vivid and memorable

(Todd, et al., 2012), influential on attentional processes (Padmala & Pessoa, 2008;

Pessoa, 2011), and associated with increased arousal (Anderson, et al., 2006).

Consequently, emotions must always be considered as a primary influence or

precursor of learning experiences (Jarvilehto, 2000a, 2001). The action tendencies,

or physical behaviour, corresponding to emergent cognition-emotion interactions will

subsequently set precedence for how an individual acts or functions in similar

situations in the future (Lewis, 1997, 2000a, 2002).

The need for a principled integration of emotion into learning

Taking into account the previous conceptualisations, the under-estimated or

even complete disregard of emotions, highlights the need for a theoretical framework

that will enable an accurate and detailed description of emotional experiences in


learning and performance (Hanin, 2007b; Headrick, et al., 2015; Vallerand &

Blanchard, 2000). Previous work investigating emotions in relation to sport has

largely focussed on a ‘snapshot’ of performance at one point in time. Studies such as

those reviewed by Lane et al. (2012) typically present emotions as either positive

(e.g. joy) or negative (e.g. fear) with respect to performance. With a broad and

complex range of emotions isolated into two categories some emotions, often

negatively orientated emotions, are considered dysfunctional for the performance of

a given task (Jones, 2003). On the other hand, so called negative emotions have also

been shown be beneficial to performance, and in some cases perceived positive

emotions have been found to be detrimental to performance (Hanin, 2007a, 2010).

The approach advocated here is that due to the dynamic emergence of emotions as

individuals interact with learning tasks, any study of emotions should consider the

learning experience holistically with affective constraints incorporated and emotions

tracked throughout (Headrick, et al., 2015; Pinder, et al., 2014). Furthermore, the

presence of emotions during learning should be embraced to create engaging and

representative learning and practice environments. Through the following sections

the aim is to meet both goals by proposing a principled model of Affective Learning

Design.

An ecological dynamics approach to emotion and learning

An ecological dynamics approach to learning recognises the mutual

relationship between an individual and the environment described through an

integration of key concepts from ecological psychology and dynamical systems

theory (Araújo, et al., 2006). In comparison, traditional approaches have focussed

primarily on the processes and structures within organisms to understand behaviour,

described as an organismic asymmetry (Davids & Araújo, 2010; Dunwoody, 2006).


Describing behaviour from the organism-environment scale takes into account how

perceptions, actions, intentions and cognitions emerge under the constraints of

information that are both internal and external to an individual (Jarvilehto, 2009;

Seifert & Davids, 2012; Shaw & Turvey, 1999; Warren, 2006). Ecological dynamics

concepts have been applied to learning and skill acquisition in relation to the

coordination of physical movement behaviour in clinical settings (Newell &

Valvano, 1998), motor development (Smith & Thelen, 2003; Thelen, 2002) and

recently in sport (Araújo, et al., 2006; Balagué, et al., 2013; Chow, et al., 2011;

Davids, et al., 2005; Vilar, et al., 2012).

Ecological psychology

A key concept in Ecological Psychology is the link between perceptual

information available in the environment and the resultant actions or behaviours of

complex system agents, such as humans (Araújo, et al., 2006). In the study of visual

perception, Gibson (1979) stated that the movements of an individual bring about

changes in information flow from which affordances (opportunities for action) are

perceived to support further movement. Therefore, information and movement are

deemed to be dependent on each other in a dynamically coupled relationship. As a

result, a cyclic process is created where action and perception feed off each other to

support goal-directed movement behaviour (Davids, et al., 2008). This concept

provides a foundation for describing how goal directed movement emerges

throughout a wide array of applications in the study of human behaviour.

Consequently, perception-action coupling is an important consideration for the

design of skill acquisition practice tasks and investigative methodologies in

experimental research.


Egon Brunswik (1956) advocated that for the study of individual-environment

relations, cues or perceptual variables should be sampled from an organism’s

environment to be representative of the environmental stimuli that they are adapted

from, and to which behaviour is intended to be generalised. The sampling of

perceptual variables from an individual’s ‘natural’ environment in an experimental

task, was captured by the term representative design (Brunswik, 1956; Dhami, et al.,

2004). More recent work has discussed the concept of representative design in

relation to the study of human performance and behaviour in contexts such as sport

(Araújo, et al., 2006; Araújo, et al., 2007). As a result, the term representative

learning design (RLD) has been adopted to highlight the importance of creating

representative learning and practicing environments where key perceptual variables

are sampled from the performance setting (Pinder, Davids, et al., 2011b). By

incorporating representative design into practice tasks, practitioners ensure that the

crucial cyclical relationship between perception and action in a performance

environment is maintained (Gibson, 1979). To this point, empirical work discussing

RLD has focussed on visual information provided in performance tasks (Pinder,

Davids, et al., 2011a), practice in differing performance environments (Barris,

Davids, et al., 2013), and changes to the complexity of tasks in team games

(Travassos, et al., 2012). Similarly, learning and development studies have

concentrated on the influence of practice on the organisation of movement tendencies

or action (Newell, et al., 2001) While these advancements have provided a solid

foundation for developing RLD, it is postulated that further progress can be made by

taking into account the role of emotions in devising learning tasks to effectively

simulate future performance.


Dynamic systems theory and emotions

An individual’s intentions, perceptions and actions in specific environments are

strongly influenced by the type and intensity of emotions experienced (Jones, 2003;

LaBar & Cabeza, 2006). This idea fits well within a complex systems approach that

considers an individual as many interacting parts that self-organise under the

influence of individual, environment, and task constraints (see Newell, 1986) to

achieve specific objectives (Kelso, 1995). Therefore in the regulation of human

behaviour, the perceptions, actions and cognitions of an individual self-organize to

form emergent physical and psychological responses (Warren, 2006).

Complex dynamic systems such as humans have the capability to self-organise

their actions or behaviour to achieve specific task objectives without direct input

from higher order structures or predetermined rules (Kelso, 1995; Lewis, 2000b).

When considering a human as a complex dynamic system, the informational

variables in an environment, along with associated goal objectives and intentions,

influence how an individual behaves in specific contexts. This takes into account the

self-organising processes of emotions, cognitions, and behaviours in relation to both

physical and psychological activity (Kelso, 1994; Lewis, 1996). States of system

organisation that are stable for a given task are described as attractors. Attractors are

considered to have deep ‘wells’ (see Figure 6.1), which represent previously learnt

and commonly adopted patterns of behaviour such as an emotional response like

excitement when playing fun games (Kelso, 1995; Renshaw, et al., 2012; Thelen &

Smith, 1994). Depending on the depth or stability of an attractor, changes in

informational variables (e.g. control parameters, see Balagué, et al., 2012; Passos, et

al., 2008) and/or task objectives have the potential to perturb stable states of

behaviour. Perturbations lead to a phase transition in the state of the system, often


producing instability and changes to order parameters. Unstable states, or repellors,

correspond to a ‘hill’ above potential ‘wells’ (see Figure 6.1) where behaviour is

highly variable and often less functional . Unstable states are easily influenced by

changes to informational variables both internal and external to the system

(individual). For example, an increase in an internal informational constraint such as

anxiety could lead to a change in the perceptions of the climbability of a rock wall.

Conversely, changes in perceptions of task constraints, such as facing a cricket

bowler with a reputation for hitting batters, could lead to an increase in anxiety and

lead to movement changes through freezing of the degrees of freedom (Bernstein,

1967).

Figure 6.1. Changes in complex system stability. (1) represents a weak attractor state within a shallow

(unstable) well. (2) represents a metastable region where the system lingers between attractor states

(i.e. 1 & 3). (3) represents a strong attractor state with a deep (stable) well

During periods of change, a previously unstable repellor may emerge into a

more stable attractor, initially as a shallow well, before potentially strengthening to a

deep well (Kelso, 1995; Kelso & Tognoli, 2009; Vallacher & Nowak, 2009).

Therefore, while initially an individual adopts a novel and functional response,

portrayed as a stable pattern of organisation, over time the pattern becomes a


characteristic response to a specific experience or environment (Lewis, 2000b;

Thelen, 2002; Thelen & Smith, 1994). In some cases an individual may develop

several coexisting attractor states that are available to choose from, referred to as

multistability (Kelso, 2012). A system displaying multistability has the ability to

switch between selected attractor states as necessary, which can be advantageous

with regards to adaptability, but also potentially detrimental due to inconsistency in

performance (Chow, et al., 2011; Kelso, 2012). A system can also display

metastability, which is a partial and temporary state or pivot where a system is poised

to emerge into a new stable state (Kelso, 2012; Kelso & Tognoli, 2009).

Metastability reflects the tendency of system components competing in an attempt to

function both individually, and in unison concurrently (Kelso, 2008). Therefore,

metastability is a state of partial organisation that allows a system to function while

transitioning (i.e. learning) between states of organisation (Chow, et al., 2011;

Pinder, et al., 2012).

Learning

Learning and performance are difficult to separate from each other, with

learning processes derived from observable performances, typically under highly

controlled conditions (i.e. learning to perform) (Schmidt & Wrisberg, 2008). As

learning is inferred from performance, learning and experimental environments

should sample the conditions of the relevant performance environment to uphold

representative learning design (Pinder, Davids, et al., 2011b). In complex systems

terminology, learning can be described as the transition to a preferred functional state

(attractor) from a previously learnt and now ineffective state (Chow, et al., 2008;

Jarvilehto, 2009; Liu, et al., 2006; Thelen, 1995). Learning and development are

often discussed alongside each other as distinct, yet integrated concepts (Newell, et


al., 2001). In general, learning relates to changes (e.g. performance, ability) over

short time scales while development describes changes over longer periods of time

(Granott & Parziale, 2002; Newell, et al., 2001). The interaction between the time

scale of perception-action (short term – seconds and minutes) and the time scale of

learning and development (long term – days, weeks, and months) influences how

situational constraints of an environment are approached (Davids, et al., 2001; Lewis,

2002; Thelen, 1995). From a dynamic systems perspective, short term events shape

long term developmental patterns or states, and conversely the established

developmental states influence performance and learning in real time (Lewis, 2000a,

2002). These reciprocal influences between short and long term experiences have

been described as ‘cascading constraints’ on learning and performance (Lewis, 1997,

2002). Therefore established EI’s of an experience influence how an individual will

approach learning an unfamiliar task (Lewis, 2000a, 2004).

In summary, ecological dynamics concepts have been adopted to describe how

individuals develop interpretations of experiences in relation to the self-organising

relationship between cognition and emotion (Lewis, 1996, 2000a, 2000b, 2002,

2004; Thelen & Smith, 1994). During development, the relationship between

cognitions and emotions in respect to an experience may change, resulting in phase

transitions from unstable to stable states of organisation (Fogel, et al., 1992; Lewis,

2004; Thelen & Smith, 1994). Like physical behaviour, the development of

cognition-emotion links may lead to metastable periods where an individual has the

potential to adopt one of a ‘cluster’ of possible states, indicating a sensitivity to

changes in situational constraints (Hollis, et al., 2009; Lewis, 2000b, 2004). During

a period of metastability, behavioural tendencies would be expected to exhibit

increased variability until a more stable state of self-organisation emerges (Chow, et


al., 2011; Hollis, et al., 2009). Accompanying this variability in behaviour, variable

and individualised emotional responses would also emerge reflecting the instability

in the system as exploration of the environment and learning takes place (de Weerth

& van Gert, 2000; Lewis, 1996, 2004).

A model of affective learning design

By capturing the theoretical concepts and processes discussed in the previous

sections, the remainder of this paper will introduce and discuss a model of affective

learning design. Affective learning design (ALD) has been posited as a

complementary aspect of representative learning design (Pinder, Davids, et al.,

2011b; Renshaw, et al., 2014). ALD is focussed on the critical interactions between

emotions, cognitions and behaviours that underpin a holistic approach to learning

(Headrick, et al., 2015; Pinder, et al., 2014). This proposed model provides a

principled description of learning founded on an ecological dynamics approach in

unison with the experiential knowledge of coaches, sport psychologists, pedagogues,

and sport scientists. The key phases of the model are evaluation, planning,

implementation, and observation & monitoring (see Figure 6.2). Together these

phases form a cyclical loop where each phase underpins the next across short and

long time scales. Due to the cyclical nature of the model there is no standardised start

or end points. Instead the model is open for interpretation from various phases.

Figure 6.2. A model of Affective Learning Design


Evaluation

The evaluation phase of the ALD model concentrates on how the capabilities,

or intrinsic dynamics of individuals are matched to the demands of a task (Newell &

Ranganathan, 2010; Schöner, et al., 1992). Assessing the convergence (‘match up’)

between intrinsic dynamics and task demands involves identifying individual rate

limiters (Thelen, 1995) on performance that must be overcome in order for the

individual to be successful when experiencing specific situational constraints

(Davids, et al., 2001). From an ALD perspective, rate limiters could be individual

task or environmental constraints. Potential individual constraints include factors

such as one, or a combination of emotions, cognitions, and actions relating to the task

that are restricting or preventing successful performance. Generic rate limiters

include lack of physical strength, poor perceptual skills, an unstable/dysfunctional

emotion-cognition state, misguided goals or intentions, and low self-efficacy. Rate

limiters such as these may be identified to be barriers to successful performance

alone, or in synchronisation with one task or environmental constraints. For

example, the interaction between balance beam performance and (dys)functional

emotions (Cottyn, et al., 2012), track sprinting success and self-efficacy (Gernigon &

Delloye, 2003), the effect of induced anxiety on movement behaviour whilst

climbing (Pijpers, et al., 2005), throwing darts, and shooting free throws in basketball

(Oudejans & Pijpers, 2009).

Identifying key rate limiters for performance is a challenge that can be

achieved through various methods. Performance outcomes (e.g. time, score),

performance analysis (number of possessions, distance travelled), biomechanical

analysis (joint angles, velocities), physiological analysis (heart rate, oxygen

consumption, sweat response), and pure observation (video, real time) are all


common techniques used to evaluate specific aspects of performance (Cooke,

Kavussanu, McIntyre, Boardley, & Ring, 2011; Glazier, 2010). However, evaluating

which measures are most meaningful and applicable to the individual and task in

question is the key challenge for a coach, or sport scientist. One rarely accessed

source of information that can be used to help identify key rate limiters is the

experiential knowledge of coaches and teachers, sport psychologists, sport scientists

to highlight aspects of performance that require improvement (Greenwood, Davids,

& Renshaw, 2014).

To ascertain subjective components of performance such as emotions and

cognitions, the coach-athlete relationship is also of great importance (Davis &

Jowett, 2014; Greenwood, Davids, & Renshaw, 2012). Determining the emotions

and cognitions associated with physical performance can be achieved through

methods including self-report measures (e.g. questionnaires, rating scales), self-

confrontation techniques (guided reviews of performance), and formal/informal

communication (interviews) (e.g. Gernigon, et al., 2004; Pijpers, et al., 2005).

In relation to ecological dynamics concepts, the evaluation phase represents an

existing stable state of system (intrinsic dynamics) organisation, which may or may

not be functional for the task. The identification of rate limiters through methods

such as those previously mentioned, indicates that the existing intrinsic dynamics of

the individual are not functional in relation to the current and/or future task demands

(Newell & Ranganathan, 2010). Therefore, as detailed in the following sections, the

system needs to be somewhat destabilised in order to facilitate a phase transition to a

different attractor state that is functional for performance (Jarvilehto, 2009; Kelso &

Zanone, 2002). The critical role of an individual and/or coach is to assess previous

performance and recognise control parameters that can be manipulated to create


metastable periods where an individual must explore for new functional states of

organisation (Kelso, 2012; Pinder, et al., 2012).

Planning

Following the evaluation of existing intrinsic dynamics–task demands

convergence and identification of key rate limiters, the next phase of the model

involves planning the design of the learning event. As with the evaluation phase, the

integration of experiential knowledge with theoretical underpinning concepts is

crucial for designing learning experiences that effectively cater for the individual

needs of a performer (Renshaw, et al., 2009). The planning phase concentrates on

identifying control parameters (informational variables linked with rate limiters) that

can be manipulated in order to create instability in performance, therefore forcing the

individual to search for novel solutions to task demands (Balagué, et al., 2012;

Warren, 2006). Through scaled changes to control parameters, a system can be

destabilised resulting in phase transitions, self-organisation, and ultimately a change

in order parameter (Passos, Araújo, & Davids, 2013; Warren, 2006). Examples of

control parameters in sport include relative velocity and interpersonal distance in

Rugby Union sub-phases (Passos, et al., 2008). Critically, control parameters do not

directly relate to specific changes in system order, but can be manipulated to promote

instability and facilitate self-organisation (Thelen, 2002). The aim of manipulating

control parameters is therefore to guide an individual into regions of metastability

(see Figure 6.1), to provide opportunities to transition between co-existing states of

organisation (Kelso, 2012).

Tasks that bring about metastable states are commonly associated with

increased variability in emotions, cognitions, and actions (Lewis, 2004). Increased

variability suggests that the individual is exploring the environment in search for new


attractors which will inherently be accompanied by periods of failure and uncertainty

before success or functionality is achieved (Alexandrov & Jarvilehto, 1993; Collins

& MacNamara, 2012; Warren, 2006). Therefore when planning a learning event it is

imperative to consider the influence that successful/unsuccessful performances might

have on psychological concepts including self-efficacy and motivation (Renshaw, et

al., 2012). From an ALD approach these psychological (individual biased) concepts

are considered in unison with physical performance outcomes and achievements

(efficacy) of an individual under specific situational constraints (Davids, et al., 2001;

Gigerenzer & Goldstein, 1996; Simon, 1990). This reflects the interaction between

the individual and the environment instead of solely focussing on the internal

thoughts and verbalisations of the performer (Davids & Araújo, 2010; Jarvilehto,

1998a; Simon, 1956). Consequently, when planning learning events a coach should

take into account thoughts and emotions expressed or displayed by a performer,

along with observed actions and performance outcomes (Headrick, et al., 2015).

Specifying clear goals and intentions for a learning event is another key aspect

of the planning phase. Designing tasks with set goals provides some context to a

learning event and subsequently a reason or motive for participation (Araújo, et al.,

2006; Shaw & Turvey, 1999). Performance goals relate to being successful, and

being judged by others as competent, while mastery goals refer to the development

of skills and competency (Ames & Archer, 1987; Elliot, 2005). Competence is one’s

mastery of a task in relation to the task demands, an individual’s own previous and

potential performance achievements, and in comparison to the achievements of

others (Elliot, 2005). In the case of performance orientated tasks, a performer would

be expected to display competence and achieve successful outcomes, indicating they

are in a stable and functional attractor state. Mastery focussed tasks on the other


hand would be expected to be associated with greater variability and reduced

functional performance outcomes, suggesting a metastable state (Pinder, et al., 2012;

van de Pol, et al., 2012a). From an ALD approach these differences in performance

incorporate actions, cognitions, and emotions to provide a holistic understanding of

how an individual interacts with the situational constraints of a learning environment

(Davids, et al., 2001; Headrick, et al., 2015). Establishing whether a future session

or task is orientated toward performance or mastery goals is therefore critical for

stipulating the purpose of the learning event.

Implementation

Implementing plans devised to address mismatches between intrinsic dynamics

and (expected) task demands is the next phase of the ALD model. While this phase

is again theoretically founded on ecological dynamics concepts, it now requires

theoretical and strategic planning to be incorporated into practice through the skills

of a coach (Greenwood, et al., 2012; Renshaw, et al., 2009; Renshaw, et al., 2012).

The planning and implementation phases are therefore closely linked and have the

potential to influence each other from one learning event to another. Central to the

implementation phase is the use of constraints to guide the self-organisation of

emotional, cognitive, and behavioural tendencies (Newell, 1986; Newell &

Ranganathan, 2010). Constraints are commonly associated with physical or

structural aspects of an environment that act as boundaries or restrictions on

behaviour. For example, the use of a ‘battle zone’ for cricket practice where the

playing area is restricted in comparison to a full game (Renshaw, Chappell, et al.,

2010), or the use of implements exhibiting different haptic characteristics during a

dynamic interceptive task (Headrick, Renshaw, et al., 2012).


From an ALD approach, informational constraints that specify situational,

contextual, instructional aspects of a task must also be taken into account (Davids, et

al., 2001; Headrick, et al., 2015; Newell & Ranganathan, 2010; Seifert & Davids,

2012). Previous applications of informational constraints to sport performance

include specifying instructions relating to game play scenarios in basketball sub-

phases (Cordovil, et al., 2009), and inducing anxiety by increasing the height of

climbing traverses (Pijpers, et al., 2006). Including informational constraints affords

the opportunity to create practice vignettes that stipulate situational and contextual

information (e.g. time, score and game scenario) to engage the emotions and

thoughts of a performer (Davids, et al., 2013; Headrick, et al., 2015). By effectively

manipulating key constraints the role of the coach is to create an environment that

facilitates learning, and the emergence of successful/functional performance, without

becoming overly prescriptive (Millar, Oldham, & Donovan, 2011; Renshaw,

Oldham, Glazier, & Davids, 2004). Therefore performers are encouraged to explore

and problem solve in environments that are designed to target and address identified

rate limiters (Renshaw, et al., 2009; Renshaw, et al., 2012).

When planning and then implementing the constraints that will be applied to

learning events, a major challenge is ensuring that the environment is representative

of the constraints experienced in a performance or match environment (RLD - see

Pinder, Davids, et al., 2011b). Designing and executing learning events to take a

holistic approach to the implementation of constraints ensures that physical (e.g.

technical), physiological and psychological aspects of performance are

accommodated for during learning and practice (Pinder, et al., 2014; Renshaw, et al.,

2004). From this perspective, learning to exploit the emotional constraints of a task

is equally as important as developing technique or physical fitness for a specific task


(Headrick, et al., 2015; Pinder, et al., 2014). In order to facilitate challenging and

engaging practice environments a coach must at times manipulate constraints with

the intention of creating periods of failure (instability) (Chow, et al., 2011; Collins &

MacNamara, 2012). Inducing failure through the disruption of behaviour that had

previously ‘worked’ or ‘felt good’ has potential psychological consequences

(Gernigon & Delloye, 2003; Pinder, et al., 2014; Thelen & Smith, 1994). Therefore,

coaches must judge carefully how representative a learning environment should be

based on their knowledge of and relationship with an individual. In order to nurture

the psychological state of performers, learning environments will rarely be

completely representative of competition conditions, but must attempt to stimulate as

many key aspects of the intended competitive setting as possible (Brunswik, 1956;

Pinder, Davids, et al., 2011b).

Observation and monitoring

Observation and monitoring is the final phase of the model of ALD. This

phase involves within session or ‘real time’ assessment of the plans implemented and

how the learning event is addressing the targeted rate limiters. Where the previously

discussed evaluation phase involves detailed analysis methods to identify key rate

limiters, the observation and monitoring phase involves more subjective methods

based on the experiential knowledge of a coach/practitioner and their understanding

of the performer (Greenwood, et al., 2014). These ‘on the run’ methods might

include the visual inspection of action tendencies, communication with the

performer, assessing group engagement in a task, or recording fundamental

performance outcomes (i.e. completion times, scores). Therefore, the

implementation phase along with this observation and monitoring phase form a real

time loop within the ALD model (see Figure 6.2). This loop provides an avenue for


on-line modifications during the learning event to deal with unexpected emotions,

cognitions or behaviours through refinements to the learning experience within the

broad objectives of the session.

A key observational indicator of performance from an ALD approach is

determining the intentions of an individual in regards to displaying competence in a

task, when compared with their own previous performances or those of others (Elliot,

2005). An individual may attempt to approach success and therefore display

competence; or aim to avoid failure to prevent the display of incompetence (Elliot &

Church, 1997; Elliot & Harackiewicz, 1996). The approach-avoidance dichotomy

has been investigated in Judo training bouts to explore the nature of goal orientations

in various situations (Gernigon, et al., 2004). Here it was found that goal

involvement states (approach-avoid) fluctuate according to the constraints of the task

and the emergent relationship between actions, thoughts, and emotions (Gernigon, et

al., 2004; Lewis, 2004).

From an ALD approach the observation and monitoring of learning events

should take into account a combination of actions, cognitions, and emotions. By

adopting this approach, observed changes in one of these concepts can be intertwined

or associated with others to provide an enhanced understanding of how individual

performers attempt to function in metastable periods. Gernigon et al., (2004, p. 592)

summarised this relationship stating that “…cognition is seen as indissociable from

action, feelings, and experience, and closely linked to ecological constraints”. For

example, the link between increased anxiety, completion time, and exploratory

movements in climbing tasks at different heights (Pijpers, et al., 2005), or the

relationship between balance beam performance and functional or dysfunctional

emotions due to increases in task difficulty (Cottyn, et al., 2012).


Further studies have investigated similar relationships when comparing

between practice and performance tasks. In target shooting, competition situations

corresponded with increased anxiety, decreased shooting accuracy, shorter quiet eye

(QE) periods, and several gun barrel kinematic differences when compared with the

practice situations (Causer, Holmes, Smith, & Williams, 2011). The benefits of

training with induced anxiety have also been studied in dart throwing and basketball

shooting tasks (Oudejans & Pijpers, 2009, 2010). Performance in these sport specific

tasks was found to be enhanced following training in induced anxiety conditions,

even though participants were observed to display increased heart rates, perceived

exertion ratings, and subjective anxiety scores which when isolated from

performance could be interpreted as detrimental responses. Similar findings were

observed in police officers who trained and performed handgun shooting under

pressure (Nieuwenhuys & Oudejans, 2011; Oudejans, 2008). These findings exhibit

how observable changes in performance outcomes and movement responses are

accompanied by emotional and cognitive tendencies that become recognised as

characteristic responses for specific situations (Lewis, 2000b; Lewis & Granic,

2000).

Following the observation and monitoring of these intertwined actions,

emotions, and cognitions a coach or practitioner can identify key aspects of

performance that need to be addressed further. Therefore the cyclic model of

affective learning design returns to the evaluation phase where the next ‘round’ of

evaluation, planning, and implementation can commence to create enhanced learning

environments. To highlight the potential practical implications of this applied model

the following section discusses the phases of evaluation, planning, implementation,


and observation / monitoring in relation to the preliminary studies presented in

Chapter 5.

Examples of ALD principles in practice

Example 1: Action-emotion relationships in a team passing game

During a team passing game (Endball – see Chapter 5A) session the task

constraint of maximum possession time was systematically manipulated with the aim

of creating metastability in actions (game events) and emotion intensity. Using this

example as a hypothetical, a coach may have identified slow ball movement as a rate

limiter during the evaluation of previous performances in a possession game such as

Netball. During the planning phase, interventions to address this rate limiter were

investigated and the task constraint of manipulating ball possession time was

identified as a candidate control parameter that may force players to re-organise

their perception-action couplings in order to dispose of the ball faster without

explicitly being instructed. Moving to the implementation phase a game situation

was created where the training environment remained largely representative of the

information present during match or competition conditions including: direct

opponents; field markings; and game context in the form of goals scored (Cordovil,

et al., 2009; Headrick, Davids, et al., 2012; Orth, et al., 2014). During the

constrained games (3, 2, and 1 second possession time limits) the actions and

emotions of the players were observed and monitored to determine whether the

interventions were going to plan. Observation and monitoring was carried out ‘in

session’ in the form of subjective perceptions of the coach. Key measures were also

collected at key periods during the session detailing game events (touches, passes

etc.), emotions (SLEQ), and perception of difficulty (PoD) for later analysis and

further evaluation.


Results from the evaluation of the game events and emotion intensity scores

revealed several findings regarding the impact of manipulating the constraint of

possession time. For example, one team of players reported consistently higher

emotion intensity scores across the session, and also happened to score more goals in

each period of play. Overall findings revealed that during the transition from 3 – 2

second possession time limits, increases in the number of goals scored linked with

elevated intensities of Enjoyment and Fulfilment SLEQ subscales along with

decreased Anger subscale intensities. When transitioning from the 2 – 1 second

possession game the number of touches, incomplete passes, and errors were found to

increase when more goals were scored. Finally when comparing the transition from

the 1 – 3 second game Enjoyment decreased when complete passes increased,

Nervousness increased in line with incomplete passes, PoD decreased as Fulfilment

increased, and Anger increased alongside elevated numbers of incomplete passes

(and vice versa).

Therefore when evaluating the effectiveness of the session a coach could

interpret these findings as suggesting that the manipulation of this specific constraint

did influence the intertwined actions and emotions of the players as a group. In

terms of the hypothetical objective of improving the speed of ball movement it

appears that the 1 second time limit for possession predicts elevated touches,

however, with the expectation of more errors and incomplete passes.

Example 2: Action, emotion, and cognitions in a representative cricket

batting task

In this example a cricket batting task was devised where delivery speed was

manipulated throughout a batting practice session. Again using a hypothetical skill

development scenario, the evaluation of previous performances in this case identified

the ability to face increasingly fast bowling as a rate limiter. In an attempt to address


this rate limiter a phase of planning identified the systematic manipulation of cricket

pitch length as a method for simulating faster deliveries while maintaining the

critical bowling action characteristics of a suitable bowler. To implement this plan

the manipulations were incorporated into a simulation representative of the first 10

overs of a one day cricket match, including contextual information regarding the

position of fielders, and the competitive relationship between batter and bowler.

Throughout the session observation and monitoring took place through a number of

methods. Actions (movement analysis) and performance outcomes (runs, delivery

speed, shot type and direction), emotion intensity (SLEQ), and cognitions (brief

confrontational interview questions) were all collected for within session and post

session analysis where appropriate. For this example the real time loop between

implementation and observation/monitoring phases was also in action, due to the

bowler not meeting the designated delivery speed in the final manipulation. When

this event occurred the decision was made to continue as planned with the session

despite an unexpected situation arising.

When evaluating the observations and measures collected throughout the

session a key finding revolving around total front foot movement was identified.

Specifically, shorter total distances covered by the front foot (both forward and back)

were found to strongly link with increases in Enjoyment scores, perception of

achievement ratings (PoA - based on interview responses of who ‘won’ each over),

and runs scored per over. These findings reveal clear relationships between actions

(foot movement), performance outcomes (runs), emotions (Enjoyment score), and

cognitions (PoA rating). Therefore a coach or sport scientist could deduce from

these findings that the batter in question was more functional for this task when he

made shorter (potentially more efficient) foot movements when facing increased


delivery speeds. As such the next planning phase may investigate tasks that

encourage the batter to explore employing shorter foot movements.

Summary

Overall this chapter has proposed a detailed theoretical background to the

concept of ALD arguing that action, emotion, and cognitions should all be integrated

in a holistic approach to the design of learning environments. In order to transfer

these conceptualisations into practice a four phase model of ALD is proposed

drawing on theoretical underpinnings and the experiential knowledge of a coach,

teacher, researcher, or practitioner. The key principle driving this model is the

constant integration and recognition of actions, emotions, and cognitions throughout

all phases supported by findings from the two practical examples. Future work

should look to further test the design of this model across a range of individuals and

different sports to enhance the representative and holistic nature of research tasks,

coaching approaches, and performance analysis methods.

Chapter 7: Epilogue 219

Chapter 7: Epilogue

220 Chapter 7: Epilogue

Introduction

The major aim of this PhD programme was to provide a principled approach

for the integration of emotion in learning. To achieve this aim a combination of

theoretical conceptualisations, methodological developments, and applied examples

have been proposed, reported, and discussed. The following sections provide an

overview of the pivotal findings and implications of this PhD, and how they might

inform future research from both theoretical and applied perspectives.

Theoretical findings and implications

Affective learning design

Through Stage 1 the theoretical position underpinning this PhD programme

was introduced and applied to the holistic study of action, emotions, and cognitions.

Reviewing and critiquing previous literature revealed the predominance of

traditional, cognitive biased (‘cold cognition’) approaches to the interaction between

actions, emotions, and cognitions. These approaches reflected an internal (to the

individual) bias in the emotion literature that did not account well (if at all) for the

interaction between an organism (individual) and environment in terms of

perceptions, actions, and intentions (Davids & Araújo, 2010; Warren, 2006). In

contrast, a small number of studies and theoretical modelling papers were found to

have advocated for a dynamic systems approach for discussing emotions. The work

of Lewis and colleagues (e.g., Lewis, 1996, 1997, 2000a, 2000b, 2002, 2004, 2005)

in particular supported the application of dynamic systems concepts such as self-

organisation, attractors, and phase transitions to describe the commensurate

interactions between cognitions and emotions across various time scales. Similar

conceptualisations by Jarvilehto and colleagues (e.g., Jarvilehto, 1998a, 1998b, 1999,


2000a, 2000b, 2001, 2009) discussed the organism – environment system and the

dynamic influence emotion has on the self-organisation of functional tendencies.

To develop these theoretical approaches to emotion further, a primary outcome

from this thesis is the introduction of the concept of Affective Learning Design

(ALD, Headrick, et al., 2015). Founded on ecological dynamics concepts and

complementary to Representative Learning Design (Pinder, Davids, et al., 2011b),

ALD promotes two key principles: (i) the design of emotion-laden learning

experiences that effectively simulate the constraints of performance environments in

sport; (ii) recognising individualised emotional and coordination tendencies that are

associated with different periods of learning (see Chapter 3). Therefore, an ALD

perspective supports a holistic approach to the intertwined relationship between

actions, emotions, and cognitions during representative learning environments that

simulate the contextual and situational demands of sport performance. The targeted

and systematic manipulation of constraints during these learning experiences aims to

force individuals into metastable periods where self-organisation (of actions

emotions, and cognitions) is required in order for functional, goal directed behaviour

to emerge (Hristovski, et al., 2011; Kelso, 1995, 2012).

The design and findings of the studies discussed in Chapter 5 vindicates this

novel theoretical approach, and its application in future theoretically informed

research. Very few previous studies have set out to investigate, let alone

demonstrate, the dynamic interaction between actions, emotions, and cognitions in

sport (with the exception of Gernigon, et al., 2004). The findings of the cricket

batting investigation in particular, reveals the complimentary interaction between

goals (i.e. game plans), emotions, and actions (outcomes, movement characteristics),

supporting the second principle of ALD above. Consequently, this thesis has


proposed an innovative, theoretically driven approach for incorporating emotion into

learning. It has also demonstrated how this approach can be applied in sport contexts,

providing an outstanding foundation for future work, with potentially ground-

breaking implications.

Affective constraints as control parameters

In reference to constraints, the theoretical underpinnings and practical findings

discussed in the following sections also provide support for the recognition of

‘affective constraints’ on emergent behaviour (Headrick, et al., 2015). Cognitive

science literature suggests that emotions are controlled or shaped by cognitions,

however the modelling of Lewis (Lewis, 2004, 2005; Lewis & Stieben, 2004; Lewis

& Todd, 2005), Jarvilehto (Jarvilehto, 2001, 2009), and findings reported in Chapters

5 and 6 indicate that emotion also influences cognition in an intertwined relationship

along with action tendencies. The development of the SLEQ (Chapter 4) for

measuring emotion intensity during learning is underpinned by the conceptualisation

of affective constraints. By considering SLEQ scores as intensities rather than

categorising responses into positive or negative valences, the opportunity was created

for individualised thresholds or regions to emerge, indicative of predicted transitions

from stable to unstable system states (Kelso, 1990, 1995). Therefore, specific

intensities of emotion can be considered as control parameters that drive a dynamic

system (individual) towards metastable regions of organisation (Kelso, 1995; Lewis,

2000b, 2004; Passos, et al., 2008). It is within these metastable regions that

increased variability in performance, intentions, and emotions would be expected as

an individual searches for functional tendencies in relation to his/her intentions

(Davids, Araújo, Seifert, & Orth, 2015; Lewis, 2004). Consequently, it is essential

that those designing learning experiences consider the sampling of contextual or


situational information alongside more physical constraints (e.g. visual information),

to account for the influence of affective constraints on emergent behaviour.

For example increased intensities of anxiety can often be functional for

learning up to a specific threshold, but beyond that become dysfunctional and

detrimental to learning due to corresponding negative influences (Jones, 1995; Kelso,

1995; Lewis, 1996). This draws parallels with the Cusp Catastrophe Theory of

Hardy and Fazey (1987), where increases in physiological arousal (or somatic

anxiety) are believed to facilitate successful performance up to a threshold, but

beyond this point a ‘catastrophe’ occurs and performance deteriorates dramatically

(Fazey & Hardy, 1988). The catastrophe point in performance is also dependant on

increased cognitive anxiety levels as part of this multi-dimensional model (Hardy,

1996; Hardy & Parfitt, 1991). However, the Catastrophe model differs from the

approach advocated throughout this thesis in that it focusses on individual (internal

to the individual) constraints, rather than the interaction between constraints shared

by the individual and environment (Jarvilehto, 2009; Lewis, 2004; Newell, 1986;

Newell, et al., 2001).

Emotions such as enjoyment could be also considered a constraint by

individuals of varying experience or expertise. For example, higher intensities of

enjoyment might be essential to engage children or novice performers in learning

tasks, whereas adults or experts who are already ‘invested’ in a sport would be

expected to exhibit more intrinsic motivation, and therefore require lower thresholds

of enjoyment to remain engaged in a task (Deci & Ryan, 1985; Renshaw, et al.,

2012; Ryan & Deci, 2000b; Toering et al., 2011). These conceptualisations

incorporate aspects of Self-Determination theory, where intrinsically motivated

individuals exhibit increased levels of competence, autonomy, and relatedness for a


specific task (Deci & Ryan, 1985, 2008; Renshaw, et al., 2012). Self-deterministic

individuals also tend to enjoy learning tasks more without the influence of external

influences, due to an innate interest or engagement in the task (Moy, Renshaw, &

Davids, in press). The incorporation and manipulation of affective constraints in

learning environments must therefore be considered from an individualised

perspective due to characteristics (organismic constraints, e.g. personality,

experience) that distinguish individual learners (Lewis, 1997).

Overall, the theoretical implications discussed highlight the novel approach to

emotion in learning advocated throughout this thesis. This represents a substantial

contribution to the literature, which was found to be lacking a principled approach to

the role of emotion in learning environments, particularly in sport contexts. The

principles of ALD therefore provide clear and theoretically sound underpinnings for

adopting these advancements in practice, as discussed in the following section.

Practical findings and implications

Development of the Sport Learning and Emotions Questionnaire (SLEQ)

Following the above theoretical conceptualisations the next stage of the PhD

programme was to provide evidence supporting the adoption of ALD principles in

practice. In order to carry out this aim the first step was to determine a method for

measuring or tracking emotion during a learning experience. Reviewing the

literature revealed a dearth of methods for measuring emotions during learning, with

existing methods considered unsuitable as they were specifically designed for pre /

post competition contexts; focused on a select few or single emotions; were static in

nature; or were originally designed for environments other than sport (e.g. PANAS,

POMS). The only questionnaire developed and currently used within sport that was

deemed to be even close to meeting the needs of the experimental studies was the


Sport Emotion Questionnaire (SEQ) of Jones et al. (2005). However, the SEQ was

only validated for use before competition. Therefore there was a clear need to

develop a new emotion measurement questionnaire specific to learning in sport. This

new questionnaire needed to be able to capture and track emotions throughout a

learning event and be as unobtrusive, and user friendly as possible. While not

ultimately suitable to the purposes of this programme of work, the SEQ provided a

framework or guide for the development of the new questionnaire. The construction

process (Chapter 4) consisted of three phases that were all completed in sport

orientated contexts. During phase 1, emotion items (words/phrases) relevant to

learning in sport were identified by survey respondents before being trimmed for

initial comparisons with those words generated during the SEQ development.

Findings from this phase revealed obvious differences in emotion items associated

with learning versus competition, and provided further justification for the

development of a new emotions measurement tool specific to learning, rather than

simply using the existing SEQ.

In phase 2 the items and the five factor structure of the SEQ were subjected to

face validation by another cohort of survey respondents to determine the relevance to

learning of each item overall, and the relevance of items to each of the SEQ factors

(i.e., subscales: anxiety; dejection; anger; excitement; happiness). Findings revealed

that, at best, only three of these factors (anxiety, excitement, and happiness) were

deemed relevant to learning, therefore warranting further investigation and

development of potentially new factors. Hence, phase 3 asked another cohort of

respondents to rate how relevant they perceived each of the remaining emotion items

to learning in sport. These results were subjected to factor analysis techniques

culminating in four learning specific factors with relevant items loading on to each.


Based on the loading items these factors were labelled as Enjoyment (5 items),

Nervousness (4 items), Fulfilment (5 items), and Anger (3 items). Structural

Equation Modelling analyses found this proposed questionnaire structure to fit or

represent the response data adequately. As a result this new proposed tool for

measuring emotion intensity during learning in sport was coined the Sport Learning

and Emotions Questionnaire (SLEQ). Considered in tandem, the conceptualisation

of ALD and development of the SLEQ represent two major contributions to the field.

The SLEQ was designed to produce a total score for emotion intensity (i.e.

engagement) along with individual scores for the four subscales. The fact that the

SLEQ records emotion intensity rather than individual scores for particular emotions

reflects a key addition to the literature (as discussed in the theoretical implications).

As mentioned previously a desired feature of this tool was the capability to track

emotion intensity across several time points to determine the influence of

manipulations to constraints, reflecting the principles of ALD. Therefore the

development of the SLEQ represents a major practical contribution of this PhD

programme to the fields of motor learning and sport and exercise psychology. Given

that the SLEQ was designed within a sport context, for use in sport contexts, it has

great potential for use by researchers, coaches, teachers, and sport practitioners

interested in how emotion impacts on skill learning tasks, sessions, or programmes.

Application of the SLEQ

Following the extensive development phases of the SLEQ, the next objective

of the PhD programme was to provide examples of how this badly needed tool could

be implemented in applied sport tasks. Chapters 5 and 6 detailed two investigations

where the SLEQ was implemented to collect emotion intensity scores alongside

action (e.g. movement characteristics, performance outcomes) and cognition (e.g.


perception of difficulty, confrontational interview responses) variables. Both

examples provided evidence to suggest that manipulating constraints to produce

metastable regions related to intertwined changes in all three categories of variables.

This indicates that individuals were forced to explore, and attempt to exploit the

variable task demands, exhibiting self-organising tendencies by adopting functional

emergent behaviour (Davids, et al., 2005; Davids, et al., 2013; Hristovski, et al.,

2006).

Findings from the Endball example (Chapter 5A) revealed that manipulations

to the task constraint of maximum ball possession time produced interlinked changes

in SLEQ score and game events. This was particularly evident during the transitions

from different games, for example, when comparing the 2nd

half of the 3 second

game with the 1st half of the 2 second game. The interlinked changes in variables

(e.g. Enjoyment and goals scored) highlighted the practical value of the SLEQ in

terms of providing emotion intensity data at the time of critical game events, and also

distinguishing the unique individual and team tendencies predicted through the

application of a dynamic systems approach to emotions in sport (Lewis, 2000b,

2004).

The second applied case example (Chapter 5B) reported on a single case

investigation using cricket batting as a task vehicle for observing interactions

between actions, emotions, and cognitions. The design of this study provided an

abundance of rich data from the SLEQ, confrontational interviews taken after each

over, along with performance outcomes (runs, shot types) and in-depth movement

variables (foot movements, bat lift height). Through the incorporation of ALD

principles the batting task included clear contextual information specifying the

situation of the simulated game and encouraging competition between the batter and


bowler. Pitch length was manipulated in order to simulate a bowler capable of

bowling faster paced deliveries, and once again the transition between different

conditions appeared to reveal the most evident links between measures. Specifically,

correlated relationships were observed between total foot movement distance,

Enjoyment subscale scores, runs, and perception of achievement. Including

confrontational questioning to determine cognitive responses provided crucial

insights into the game plan or intentions of the batter. On several occasions the

intentions of the batter were reflected in the observable performance outcomes,

movement characteristics and emotion intensities. These findings further supported

the intertwined nature of the relationship between, actions, emotions, and cognitions,

reinforcing the importance of considering all such variables in unison during applied

sport tasks.

A second key practical implication from this cricket example was the

effectiveness of shortening pitch lengths to manipulate the space-time demands of a

cricket batting task. The findings of this study revealed that repositioning the

bowler, along with key vertical reference information (e.g. stumps, umpire), was

successful in simulating targeted increases in delivery speed. This was evidenced by

the observed fluctuations in performance characteristics such as shot types, runs

scored, and foot movement changes as observed in previous work (Davids, et al.,

2015; Pinder, et al., 2012). Recording emotion intensity, thoughts and intentions

throughout the targeted speed manipulations provided further support for this method

given that the batter reported noticeable changes in these variables. Therefore this

method is proposed as a novel approach for manipulating delivery speed and

retaining key bowling action characteristics that would otherwise be lost when using

a ball projection machine (Pinder, 2012; Pinder, Davids, et al., 2011b; Pinder,


Renshaw, et al., 2011). Given that cricket coaches are often limited in terms of the

availability of bowlers, the implementation of targeted constraints to replicate

different delivery speeds has great potential for adoption into cricket (and other

interceptive action sport) talent development programmes.

Support for case study approaches

Integrating the theoretical implications, SLEQ development, and findings from

the two applied studies enables this PhD programme to demonstrate strong support

for adopting individualised case study approaches to the study of actions, emotions,

and cognitions across various time scales of learning. Findings such as the range of

SLEQ scores between individuals in the Endball study, and the individualised

movement characteristics of the cricket batter indicate that individual tendencies are

an integral part of skill acquisition research and practice. Therefore, in order to

appropriately conceptualise a learner as a complex system all contributing aspects of

emergent behaviour must be considered, observed, and evaluated. Based on

noticeable fluctuations in key measures following the manipulation of constraints

(e.g. foot movement and Enjoyment in the cricket study), case study designs provide

the opportunity to develop in-depth knowledge about the individual. This individual

specific knowledge can be used to inform the design of tailored learning

environments, as predicated in the model of ALD discussed in the following section.

Hence the methodological approaches described in Chapter 5 provide a pivotal

framework for coaches, and sport practitioners to incorporate in applied sport

settings going forward.

A conceptual model of ALD

Chapter 6 of the PhD programme brings the vital implications of the previous

chapters together to propose a model of ALD. This model is founded on the


theoretical framework of ALD and informed by the experiential knowledge of a

coach or sport practitioner. Four interlinked phases (evaluation, planning,

implementation, and observation/monitoring) are discussed, specifying both

theoretical interpretations and practical considerations. The two applied case

examples from Chapter 5 are then revisited and discussed within the framework of

the model to demonstrate how ALD principles and the SLEQ can be integrated

across different interacting time scales of learning (Lewis, 2002; Newell, et al.,

2001). Time scales of learning are reflected in the design of the model in terms of

the overall cyclical structure where previous implementations and evaluations inform

subsequent planning phases (cascading constraints, see Lewis, 1997). The ‘real time

loop’ also represents a dynamic process of constantly updating the constraints of a

task in response to changes in actions, emotions, or cognitions, indicating the task is

no longer addressing key rate limiters identified in earlier phases.

A key feature of this model to be considered by coaches and sport practitioners

is the focus on incorporating affective constraints to create representative learning

designs that immerse learners in situations that sample information from the

simulated performance environment (Pinder, et al., 2015; Pinder, et al., 2013; Pinder,

et al., 2014). Therefore, learners are constantly exposed to crucial perceptual

variables that enhance aspects such as task engagement, relatedness, enjoyment, and

investment, underpinning nonlinear pedagogy and game - based approaches to

learning (Chow, et al., 2015; Chow, et al., 2006; Renshaw, Chow, et al., 2010).

Taking these key features into account, this model is readily accessible to applied

practitioners and researchers alike and provides a comprehensive summary of the

novel approaches advocated throughout the PhD programme.



Limitations of this PhD programme that should be acknowledged focus largely

on the experimental studies described in Chapter 5. The small sample size in each of

these studies somewhat limits the conclusions able to be drawn from the findings.

However, as described previously, a key feature of the cricket batting study in

particular was the in-depth single case design that provided individual specific

manipulations and emergent behaviours to be analysed (Barker, Mellalieu,

McCarthy, Jones, & Moran, 2013). Nevertheless, future work should look to recruit

increased numbers of participants to take part in the same game based session to

identify further individualised trends. This could include batters of different skill

levels and bowlers from different ranges of standard delivery speeds (both faster and

slower) to simulate a range of game situations (e.g. overs 30 – 40). Furthermore, in

subsequent simulated competitions the bowler should also be studied to a similar

extent to ascertain his/her emotion intensity and cognitions in contrast with the

batter, thus providing a more comprehensive interpretation of the interpersonal

interactions during the ‘game’.

The second limitation of the studies described previously is the short time

scales of analysis for observing the interaction between actions, emotions, and

cognitions. Future investigations should aim to track individuals over extended time

periods to observe how variables observed across interacting time scales (i.e.

individual sessions within a weekly programme) might fluctuate in relation to the

manipulation of targeted constraints. This would also provide the opportunity to

further assess the validity and reliability of the SLEQ with (i) increased numbers of

participants, and (ii) several time points. The implementation of the SLEQ might

also be improved through the development of an electronic version that automatically


calculates intensity scores during a session, providing immediate and crucial

information to a coach, practitioner, or performer during a task or session.

Crucially, future work should focus on recognising the influence of affective

constraints during learning, and the intertwined relationship between actions (e.g.

individual movement tendencies, performance outcomes), emotion (intensity via

SLEQ), and cognitions (e.g. intentions, strategy, goals) in a range of different sports

and simulated scenarios. Therefore, the deliberate manipulation of targeted control

parameters can be studied from a multi-faceted approach to identify the intensity or

value of such variables that predicts the emergence of metastable regions. When a

metastable region is identified, another challenge for future work is establishing how

long an individual should remain in this region for effective learning to take place,

again linking with incorporating different time scales of learning (Newell, et al.,

2001). Furthermore, the systematic manipulation of control parameters by a

researcher or coach should be compared with giving the individual the choice

(autonomy) of when to change these key variables. The latter method would be

expected to reflect different individual constraints (e.g. experience, personality),

providing additional support for an individualised or case study approach to learning

tasks.

Finally, future work should aim to adopt individualised approaches to

investigate the intertwined relationship of actions, emotions, and cognitions between

learning and performance tasks (i.e. convergence between intrinsic dynamic and task

demands, Chapter 6). Given that one of the underpinning concepts of this PhD is the

design of learning experiences that represent performance environments (i.e.

competition, match), a critical next step is analysing key variables in both situations,

for the same individual(s). Consequently, results would provide an understanding of


how closely a learning task simulates the demands of an actual match or competitive

event. This would provide evidence to support, or refute, various task designs,

depending on how effectively the key perceptual and informational variables of a

simulated, and intended performance environment are sampled (including affective

constraints).

Conclusion

In summary, this PhD programme has conceived a principled approach

advocating for the consideration and recognition of emotion during learning events in

sport. This Affective Learning Design (ALD) approach has then informed the

development of a new tool designed specifically for use in sport, the Sport Learning

and Emotions Questionnaire (SLEQ). By incorporating the SLEQ and principles of

ALD in two applied sport investigations, pivotal findings have demonstrated the

effectiveness, and value of the holistic approach advocated throughout this PhD for

understanding dynamic emergent behaviour. Conclusions drawn from these findings

have identified numerous implications for practitioners in terms of designing,

evaluating, and enhancing learning tasks in the future.

Bibliography 235

Bibliography

Abelson, R. P. (1963). Computer simulation of "hot cognition". In S. Tomkins & S.

Messick (Eds.), Computer Simulation of Personality (pp. 277-298). New

York: Wiley.

Abernethy, B., Côté, J., & Baker, J. (2002). Expert decision-making in team sports.

Canberra, ACT: Technical Report for the Australian Institute of Sport.

Ajzen, I. (1989). Attitude structure and behaviour. In A. R. Pratkanis, S. J. Breckler

& A. G. Greenwald (Eds.), Attitude Structure and Function (pp. 241-274).

Hillsdale, NJ: Lawrence Erlbaum Associates Inc.

Alexandrov, Y., & Jarvilehto, T. (1993). Activity versus reactivity in psychology and

neurophysiology. Ecological Psychology, 5, 85-103. doi:

10.1207/s15326969eco0501_4

Allen, M. S., Jones, M. V., & Sheffield, D. (2009). Causal attribution and emotion in

the days following competition. Journal of Sports Sciences, 27(5), 461-468.

doi: 10.1080/02640410802538754

Ames, C., & Archer, J. (1987). Mothers' beliefs about the role of ability and effort in

school learning. Journal of Educational Psychology, 79(4), 409-414.

Anderson, A. K., Yamaguchi, Y., Grabski, W., & Lacka, D. (2006). Emotional

memories are not all created equal: Evidence for selective memory

enhancement. Learning & Memory, 13, 711-718. doi: 10.1101/lm.388906

Anokhin, P. K. (1974). Biology and Neurophysiology of conditioned reflex and its

role in adaptive behavior. Oxford: Pergamon.

Araújo, D., & Davids, K. (2011). What exactly is acquired during skill acquisition.

Journal of Consciousness Studies, 18(3-4), 7-23.

Araújo, D., Davids, K., Bennett, S. J., Button, C., & Chapman, G. (2004). Emergence

of sport skills under constraints. In A. M. Williams & N. J. Hodges (Eds.),

Skill Acquisition in Sport: Research, Theory and Practice (pp. 409-434). New

York: Routledge.

Araújo, D., Davids, K., & Hristovski, R. (2006). The ecological dynamics of decision

making in sport. Psychology of Sport and Exercise, 7, 653-676. doi:

10.1016/j.psychsport.2006.07.002

Araújo, D., Davids, K., & Passos, P. (2007). Ecological validity, representative

design, and correspondance between experimental task constraints and

behavioral setting: Comment on Rogers, Kadar, and Costall (2005).

Ecological Psychology, 19, 69-78.

Arbuckle, J. L. (2013). IBM SPSS Amos 22 User's Guide. Armonk, NY: IBM Corp.

Atkinson, J. W. (1957). Motivational determinants of risk-taking behaviour.

Psychological Review, 64(359-372).

Balagué, N., Hristovski, R., Aragonés, D., & Tenenbaum, G. (2012). Nonlinear

model of attention focus during accumulated effort. Psychology of Sport and

Exercise, 13, 591-591.

Balagué, N., Torrents, C., Hristovski, R., Davids, K., & Araújo, D. (2013). Overview

of complex systems in sport. Journal of Systems Science and Complexity, 26,

4-13. doi: 10.1007/s11424-013-2285-0

Barker, J. B., Mellalieu, S. D., McCarthy, P. J., Jones, M. V., & Moran, A. (2013). A

review of single-case research in sport psychology 1997-2012: Research

trends and future directions. Journal of Applied Sport Psychology, 25, 4-32.

doi: 10.1080/10413200.2012.709579

236 Bibliography

Barris, S., Davids, K., & Farrow, D. (2013). Representative Learning Design in

Springboard Diving: Is dry-land training representative of a pool dive?

European Journal of Sport Science, 13(6), 638-645. doi:

10.1080/17461391.2013.770923

Barris, S., Farrow, D., & Davids, K. (2013). Do the kinematics of a baulked take-off

in springboard diving differ from those of a completed dive. Journal of Sports

Sciences, 31(3), 305-313. doi: 10.1080/02640414.2012.733018

Bartlett, R. (2007). Introduction to sports biomechanics: Analysing human movement

patterns (2nd ed.). London: Routledge.

Beatty, G. F., Fawver, B., Hancock, G. M., & Janelle, C. M. (2014). Regulating

emotions uniquely modifies reaction time, rate of force production, and

accuracy of a goal-directed motor action. Human Movement Science, 33, 1-

13. doi: 10.1016/j.humov.2013.12.001

Becker, M. W. (2012). Negative emotional photographs are identified more slowly

than positive photographs. Attention, Perception & Psychophysics, 74(6),

1241-1251. doi: 10.3758/s13414-012-0308-x

Beedie, C. J., Terry, P. C., & Lane, A. M. (2005). Distinctions between emotion and

mood. Cognition and Emotion, 19(6), 847-878.

Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological

Bulletin, 107(2), 238-246. doi: 10.1037/0033-2909.107.2.238

Bentler, P. M. (2007). On tests and indicies for evaluating structural models.

Personality and Individual Differences, 42, 825-829. doi:

10.1016/j.paid.2006.09.024

Bentley, A. F. (1941). The human skin: Philosophy's last line of defense. Philosophy

of Science, 8, 1-19.

Bernstein, N. A. (1967). The control and regulation of movements. London:

Pergamon Press.

Berthouze, L., & Lungarella, M. (2004). Motor skill acquisition under environmental

perturbations: On the necessity of alternate freezing and freeing of degrees of

freedom. Adaptive Behavior, 12, 47-64.

Bhalla, M., & Proffitt, D. R. (1999). Visual-motor recalibration in geographical slant

perception. Journal of Experimental Psychology, 25(4), 1076-1096. doi:

10.1037/0096-1523.25.4.1076

Blanchard, D. C., & Blanchard, R. J. (1988). Ethoexperimental approaches to the

biology of emotion. Annual Review of Psychology, 39, 43-68.

Bloom, B. S. (1956). Taxonomy of Educational Objectives: The Classification of

Educational Goals. London: Longman Group.

Bloom, B. S. (Ed.). (1985). Developing talent in young people. New York, NY:

Ballantine.

Bollen, K. A. (1989). Structural Equations with Latent Variables. New York, NY:

Wiley.

Borkovek, T. D. (1976). Physiological and cognitive processes in the regulation of

arousal. In G. E. Shwartz & D. Shapiro (Eds.), Consciousness and self-

regulation: Advances in research (pp. 261-312). New York, NY: Plenum.

Bower, G. H. (1981). Mood and memory. American Psychologist, 36(2), 129-148.

doi: 10.1037/0003-066X.36.2.129

Breckler, S. J. (1984). Empirical validation of affect, behaviour, and cognition as

distinct components of attitude. Journal of Personality and Social

Psychology, 47(6), 1191-1205. doi: 10.1037/0022-3514.47.6.1191

Bibliography 237

Browne, M. W., & Cudeck, R. (1993). Alternate ways of assessing model fit. In K.

A. Bollen & J. S. Long (Eds.), Testing Structural Equation Models (pp. 132-

162). Newbury Park, CA: Sage.

Bruineberg, J., & Rietveld, E. (2014). Self-ogranization, free energy minimization,

and optimal grip on a field of affordances. Frontiers in Human Neuroscience,

8, article 599. doi: 10.3389/fnhum.2014.00599

Brunswik, E. (1955). Representative design and probabilistic theory in a functional

psychology. Psychological Review, 62(3), 193-217.

Brunswik, E. (1956). Perception and the representative design of psychological

experiments (2nd ed.). Berkeley: University of California Press.

Brymer, E., & Oades, L. G. (2009). Extreme sports: A positive transformation in

courage and humility. Journal of Humanistic Psychology, 49, 114-126. doi:

10.1177/0022167808326199

Brymer, E., & Schweitzer, R. (2013). Extreme sports are good for your health: A

phenomenological understanding of fear and anxiety in extreme sport.

Journal of Health Psychology, 18(4), 477-487. doi:

10.1177/1359105312446770

Burgdorf, J., & Panksepp, J. (2006). The neurobiology of positive emotions.

Neuroscience & Biobehavioral Reviews, 30, 173-187.

Burton, D., & Naylor, S. (1997). Is anxiety really facilitative? Reaction to the myth

that cognitive anxiety always impairs sport performance. Journal of Applied

Sport Psychology, 9(2), 295-202. doi: 10.1080/10413209708406488

Buszard, T., Farrow, D., Reid, M., & Masters, R. S. W. (2014). Modifying

equipment in early skill development: A tennis perspective. Research

Quarterly for Exercise and Sport, 85(2), 218-225. doi:

10.1080/02701367.2014.893054

Byrne, B. A. (2010). Structural Equation Modelling with AMOS: Basic Concepts,

Applications, and Programming (2nd ed.). New York, NY: Routledge.

Campo, M., Mellalieu, S. D., Ferrand, C., Martinent, G., & Rosnet, E. (2012).

Emotions in team contact sports: A systematic review. The Sport

Psychologist, 26, 62-97.

Carver, C. S., & Scheier, M. F. (1990). Origins and functions of positive and

negative affect: A control-process view. Psychological Review, 97, 19-35.

doi: 10.1037/0033-295X.97.1.19

Carver, C. S., & Scheier, M. F. (1998). On the self-regulation of behavior.

Cambridge, UK: Cambridge University Press.

Carver, C. S., & Scheier, M. F. (2000). Autonomy and self-regulation. Psychological

Inquiry, 11(4), 284-291.

Carver, C. S., Sutton, S. K., & Scheier, M. F. (2000). Action, emotion, and

personality: Emerging conceptual integration. Personality and Social

Psychology Bulletin, 26, 741-751. doi: 10.1177/0146167200268008

Causer, J., Holmes, P. S., Smith, N. C., & Williams, A. M. (2011). Anxiety,

movement kinematics, and visual attention in elite-level performers. Emotion,

11(3), 595-602. doi: 10.1037/a0023225

Charness, N., Tuffiash, M., & Jastrzembski, T. (2004). Motivation, emotion, and

expert skill acquisition. In D. Y. Dai & R. J. Sternberg (Eds.), Motivation,

Emotion, and Cognition: Intergrative Perspectives on Intellectual

Functioning and Development (pp. 299-319). Mahwah, NJ: Lawrence

Erlbaum Associates.

238 Bibliography

Chase, M. A., Magyar, T. M., & Drake, B. M. (2005). Fear of injury in gymnastics:

Self-efficacy and psychological strategies to keep on tumbling. Journal of

Sports Sciences, 23(5), 465-475. doi: 10.1080/02640410400021427

Chow, J. Y., Davids, K., Button, C., & Rein, R. (2008). Dynamics of movement

patterning in learning a discrete multiarticular action. Motor Control, 12(3),

219-240.

Chow, J. Y., Davids, K., Button, C., & Renshaw, I. (2015). Nonlinear Pedagogy in

Skill Acquisition: An Introduction. London: Routledge.

Chow, J. Y., Davids, K., Button, C., Shuttleworth, R., Renshaw, I., & Araújo, D.

(2006). Nonlinear pedagogy: A constraints-led framework for understanding

emergence of game play and movement skills. Nonlinear Dynamics,

Psychology, and Life Sciences, 10, 71-103.

Chow, J. Y., Davids, K., Button, C., Shuttleworth, R., Renshaw, I., & Araújo, D.

(2007). The role of nonlinear pedagogy in physical education. Review of

Educational Research, 77(3), 251-278.

Chow, J. Y., Davids, K., Hristovski, R., Araújo, D., & Passos, P. (2011). Nonlinear

pedagogy: Learning design for self-organizing neurobiological systems. New

Ideas in Psychology, 29, 189-200. doi: 10.1016/j.newideapsych.2010.10.001

Chow, J. Y., Davids, K., Renshaw, I., & Button, C. (2013). The Acquisition of

Movement Skill in Children through Nonlinear Pedagogy. In J. Côté & R.

Lidor (Eds.), Conditions of Children's Talent Development in Sport (pp. 41-

59). Morgantown, WV: FIT.

Collins, D., & MacNamara, A. (2012). The rocky road to the top: Why talent needs

trauma. Sports Medicine, 42(11), 907-914. doi: 10.2165/11635140

Cooke, A., Kavussanu, M., McIntyre, D., Boardley, I. D., & Ring, C. (2011). Effects

of competitive pressure on expert performance: Underlying psychological,

physiological, and kinematic mechanisms. Psychophysiology, 48(8), 1146-

1156. doi: 10.1111/j.1469-8986.2011.01175.x

Cordovil, R., Araújo, D., Davids, K., Gouveia, L., Barreiros, J., Fernandes, O., &

Serpa, S. (2009). The influence of instructions and body-scaling as

constraints on decision-making processes in team sports. European Journal

of Sport Science, 9(3), 169-179.

Côté, J., Baker, J., & Abernethy, B. (2003). From play to practice: A developmental

framework for acquisition of expertise in team sports. In J. L. Starkes & K. A.

Ericsson (Eds.), Expert Performance in Sports: Advances in Research on

Sport Expertise (pp. 89-113). Champaign, IL: Human Kinetics.

Cottyn, J., De Clercq, D., Crombez, G., & Lenoir, M. (2012). The interaction of

functional and dysfunctional emotions during balance beam performance.

Research Quarterly for Exercise and Sport, 83(2), 300-307.

Cowan, E. (2011). In the firing line: diary of a season. Sydney: NewSouth

Publishing.

Cromwell, H. C., & Panksepp, J. (2011). Rethinking the cognitive revolution from a

neural perspective: How overuse/misuse of the term 'cognition' and the

neglect of affective controls in behavioral neuroscience could be delaying

progress in understanding BrainMind. Neuroscience & Biobehavioral

Reviews, 35, 2026-2035.

Curran, P. J., West, S. G., & Finch, J. F. (1996). The robustness of test statistics to

nonnormality and specfication error in confirmatory factor analysis.

Psychological Methods, 1, 16-29. doi: 10.1037/1082-989X.1.1.16

Bibliography 239

Davids, K. (2001). Games, training, and other constraints on individual performance:

A role for dynamical systems theory? Sportscience, 5(2), 1-3.

Davids, K. (2012). Learning design for nonlinear dynamical movement systems. The

Open Sports Sciences Journal, 5(Suppl 1), 9-16. doi:

10.2174/1875399X01205010009

Davids, K., & Araújo, D. (2010). The concept of 'Organismic Asymmetry' in sport

science. Journal of Science and Medicine in Sport, 13(6), 633-640. doi:

10.1016/j.jsams.2010.05.002

Davids, K., Araújo, D., Hristovski, R., Passos, P., & Chow, J. Y. (2012). Ecological

dynamics and motor learning design in sport. In N. J. Hodges & A. M.

Williams (Eds.), Skill Acquisition in Sport: Research, Theory and Practice

(2nd ed., pp. 112-130). London: Routledge.

Davids, K., Araújo, D., Seifert, L., & Orth, D. (2015). Expert performance in sport:

An ecological dynamics perspective. In D. Farrow & J. Baker (Eds.),

Routledge Handbook of Sport Expertise (pp. 130-144). London: Routledge.

Davids, K., Araújo, D., & Shuttleworth, R. (2005). Applications of dynamical

systems theory to football. In T. Reilly, J. Cabri & D. Araújo (Eds.), Science

and Football V: The Proceedings of the Fifth World Congress on Science and

Football. (pp. 537-550). Lisbon, Portugal: Routledge.

Davids, K., Araújo, D., Vilar, L., Renshaw, I., & Pinder, R. A. (2013). An ecological

dynamics approach to skill acquisition: Implications for development of

talent in sport. Talent Development & Excellence, 5, 21-34.

Davids, K., Brymer, E., Seifert, L., & Orth, D. (2014). A constraints-based approach

to the acquisition of expertise in outdoor adventure sports. In K. Davids, R.

Hristovski, D. Araújo, N. Balagué Serre, C. Button & P. Passos (Eds.),

Complex Systems in Sport (pp. 306-318). New York, NY: Routledge.

Davids, K., Button, C., & Bennett, S. (2008). Dynamics of skill acquisition: A

constraints-led approach. Champaign: Human Kinetics.

Davids, K., Glazier, P., Araújo, D., & Bartlett, R. (2003). Movement systems as

dynamical systems: The functional role of variability and it's implications for

sports medicine. Sports Medicine, 33(4), 245-260.

Davids, K., Williams, A. M., Button, C., & Court, M. (2001). An integrative

modelling approach to the study of intentional movement behavior. In R. N.

Singer, H. A. Hausenblas & C. M. Janelle (Eds.), Handbook of Sport

Psychology (2nd ed., pp. 144-173). New York, NY: John Wiley & Sons.

Davidson, R. J., & Schwartz, G. E. (1976). The psychobiology of relaxation and

related states: A multi-process theory. In D. I. Motofsky (Ed.), Behvior

control and modification of physiological activity. Englewoord Cliffs, NJ:

Prentice-Hall.

Davis, L., & Jowett, S. (2014). Coach-athlete attachment and the quality of the

coach-athlete relationship: Implications for athlete's well-being. Journal of

Sports Sciences, 32(15), 1454-1464. doi: 10.1080/02640414.2014.898183

de Weerth, C., & van Gert, P. (2000). The dynamics of emotion-related behaviors in

infancy. In M. D. Lewis & I. Granic (Eds.), Emotion, Development, and Self-

Organization: Dynamic Systems Approaches to Emotional Development (pp.

324-348). Cambridge, UK: Cambridge University Press.

Deci, E. L. (1971). Effects of externally mediated rewards on intrinsic motivation.

Journal of Personality and Social Psychology, 18, 105-115.

Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in

human behaviour. New York: Plenum.

240 Bibliography

Deci, E. L., & Ryan, R. M. (1991). A motivational approach to self: Integration in

personality. In R. Dienstbier (Ed.), Nebraska Symposium on Motivation (Vol.

38, pp. 237-288). Lincoln, NE: University of Nebraska Press.

Deci, E. L., & Ryan, R. M. (2008). Self-Determination Theory: A macrotheory of

human motivation, development, and health. Canadian Psychology, 49(3),

182-185. doi: 10.1037/a0012801

Dewar, A. J., & Kavussanu, M. (2012). Achievement goals and emotions in team

sport athletes. Sport, Exercise and Performance Psychology, 1(4), 254-267.

doi: 10.1037/a0028692

Dewey, J. (1895). The theory of emotion. (II) The significance of emotions.

Psychological Review, 2, 13-32. doi: 10.1037/h0070927

Dhami, M. K., Hertwig, R., & Hoffrage, U. (2004). The role of representative design

in an ecological approach to cognition. Psychological Bulletin, 130(6), 959-

988.

Dicks, M., Button, C., & Davids, K. (2010). Examination of gaze behaviours under

in situ and video simulation task constraints reveals differences in

information pickup for perception and action. Attention, Perception &

Psychophysics, 72(3), 706-720. doi: 10.3758/APP.72.3.706

Diener, C. I., & Dweck, C. S. (1978). An analysis of learned helplessness:

Continuous changes in performance, strategy, and achievement cognitions

following failure. Journal of Personality and Social Psychology, 36(5), 451-

462.

Diener, C. I., & Dweck, C. S. (1980). An analysis of learned helplessness: II. The

processing of success. Journal of Personality and Social Psychology, 39(5),

940-952.

Diener, E., Larsen, R. J., Levine, S., & Emmons, R. A. (1985). Intensity and

frequency: Dimensions underlying positive and negative affect. Journal of

Personality and Social Psychology, 48(5), 1253-1265. doi: 10.1037/0022-

3514.48.5.1253

Dolan, R. J. (2002). Emotion, cognition, and behaviour. Science, 298, 1191-1194.

doi: 10.1126/science.1076358

Dunwoody, P. T. (2006). The neglect of the environment by cognitive psychology.

Journal of Theoretical and Philosophical Psychology, 26, 139-153.

Dweck, C. S. (1986). Motivational processes affecting learning. American

Psychologist, 41(10), 1040-1048.

Dweck, C. S., & Elliott, E. S. (1983). Achievement motivation. In E. M.

Heatherington (Ed.), Handbook of child psychology: Socialization,

personality, and social development (Vol. 4, pp. 643-691). New York: Wiley.

Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to motivation

and personality. Psychological Review, 95(2), 256-273.

Easterbrook, J. A. (1959). The effect of emotion on cue utilization and the

organization of behaviour. Psychological Review, 66(3), 183-201. doi:

10.1037/h0047707

Edelman, G. M., & Gally, J. A. (2001). Degeneracy and complexity in biological

systems. Proceedings of the National Academy of Sciences of the United

States of America, 98(24), 13763-13768. doi: 10.1073/pnas.231499798

Egiazaryan, G. G., & Konstantin, V. S. (2007). Theory of functional systems in the

scientific school of P.K. Anokhin. Journal of the History of Neurosciences,

16, 194-207. doi: 10.1080/09647040600602805

Bibliography 241

Elliot, A. J. (2005). A conceptual history of the achievement goal construct. In A. J.

Elliot & C. S. Dweck (Eds.), Handbook of competence and motivation. New

York: The Guilford press.

Elliot, A. J., & Church, M. A. (1997). A hierarchical model of approach and

avoidance achievement motivation. Journal of Personality and Social

Psychology, 72, 218-232.

Elliot, A. J., & Harackiewicz, J. M. (1996). Approach and avoidance achievement

goals and intrinsic motivation: A mediational analysis. Journal of Personality

and Social Psychology, 70(3), 461-475.

Elliott, E. S., & Dweck, C. S. (1988). Goals: An approach to motivation and

achievement. Journal of Personality and Social Psychology, 54, 5-12.

Fazey, J. A., & Hardy, L. (1988). The inverted-U hypothesis: A catastrophe for sport

psychology. In British Association of Sports Sciences Monograph No. 1.

Leeds, United Kingdom: The National Coaching Foundation.

Field, A. (2013). Discovering statistics using SPSS (4th ed.). London: SAGE

publications.

Field, A., & Hole, G. (2003). How to design and report experiments. London: SAGE

Publications.

Fingelkurts, A. A., & Fingelkurts, A., A. (2004). Making complexity simpler:

Multivariability and metastability in the brain. International Journal of

Neuroscience, 114, 843-862. doi: 10.1080/00207450490450046

Fogel, A., Nwokah, E., Dedo, J. Y., Messinger, D., Dickson, K. L., Matusov, E., &

Holt, S. A. (1992). Social process theory of emotion: A dynamic systems

approach. Social Development, 1(2), 122-142.

Folkman, S. (2008). The case for positive emotions in the stress process. Anxiety,

Stress & Coping: An International Journal, 21, 3-14. doi:

10.1080/10615800701740457

Fredrickson, B. L., & Branigan, C. (2005). Positive emotions broaden the scope of

attention and thought-action repertoires. Cognition & Emotion, 19(3), 313-

332. doi: 10.1080/02699930441000238

Fredrickson, B. L., & Losada, M. F. (2005). Positive affect and the complex

dynamics of human flourishing. American Psychologist, 60(7), 678-686.

Freeman, W. J. (2000). Emotion is essential to all intentional behaviours. In M. D.

Lewis & I. Granic (Eds.), Emotion, Development, and Self-Organization:

Dynamic Systems Approaches to Emotional Development (pp. 209-235).


Friesen, A. P., Lane, A. M., Devenport, T. J., Sellars, C. N., Stanley, D. N., &

Beedie, C. J. (2013). Emotion in sport: considering interpersonal regularion

strategies. International Review of Sport and Exercise Psychology, 6, 139-

154. doi: 10.1080/1750984X.2012.742921

Frijda, N. H. (1986). The Emotions. Cambridge: Cambridge University Press.

Frijda, N. H. (1993). The place of appraisal in emotion. Cognition & Emotion, 7(3-

4), 357-387. doi: 10.1080/02699939308409193

Frijda, N. H. (1994). Emotions are functional, most of the time. In P. Ekman & R. J.

Davidson (Eds.), The Nature of Emotion: Fundamental Questions (pp. 112-

122). New York, NY: Oxford University Press.

Gernigon, C., d'Arripe-Longueville, F., Delignieres, D., & Ninot, G. (2004). A

dynamical systems perspective on goal involvement states in sport. Journal of

Sport & Exercise Psychology, 26, 572-596.

242 Bibliography

Gernigon, C., & Delloye, J. (2003). Self-efficacy, causal attribution, and track

athletic performance following unexpected success or failure among elite

sprinters. The Sport Psychologist, 17, 55-76.

Gibson, J. J. (1979). The ecological approach to visual perception. Hillsdale, NJ:

Erlbaum.

Gigerenzer, G., & Goldstein, D. G. (1996). Reasoning the fast and frugal way:

Models of bounded rationality. Psychological Review, 104(4), 650-669. doi:

10.1037/0033-295X.103.4.650

Glazier, S. P. (2010). Game, set and match? Substantive issues and future directions

in performance analysis. Sports Medicine, 40(8), 625-634. doi:

10.2165/11534970

Gould, D., & Krane, V. (1992). The arousal-athletic performance relationship:

Current status and future directions. In T. S. Horn (Ed.), Advances in Sport

Psychology (pp. 119-142). Champaign, IL: Human Kinetics.

Granott, N., & Parziale, J. (2002). Microdevelopment: A process-orientated

perspective for studying development and learning. In N. Granott & J.

Parziale (Eds.), Microdevelopment: Transition Processes in Development and

Learning (pp. 1-28). Cambridge: Cambridge University Press.

Graydon, M. M., Linkenauger, S. A., Teachman, B. A., & Proffitt, D. R. (2012).

Scared stiff: The influence of anxiety on the perception of action capabilities.

Cognition and Emotion, 26(7), 1301-1315. doi:

10.1080/02699931.2012.667391

Greenwood, D., Davids, K., & Renshaw, I. (2012). How elite coaches' experimental

knowledge might enhance empirical research on sport performance.

International Journal of Sports Science & Coaching, 7(2), 411-422. doi:

10.1260/1747-9541.7.2.411

Greenwood, D., Davids, K., & Renshaw, I. (2014). Experiential knowledge of expert

coaches can help identify informational constraints on performance of

dynamic interceptive actions. Journal of Sports Sciences, 32(4), 328-335. doi:

10.1080/02640414.2013.824599

Gruber, J., Kogan, A., Quoidbach, J., & Mauss, I. B. (2013). Happiness is best kept

stable: Positive emotion variability is associated with poorer psychological

health. Emotion, 13, 1-6. doi: 10.1037/a0030262

Guadagnoli, M. A., & Lee, T. D. (2004). Challenge point: A framework for

conceptualizing the effects of various practice conditions in motor learning.

Journal of Motor Behavior, 36(2), 212-224. doi: 10.3200/JMBR.36.2.212-

224

Hammond, K. (1998). Ecological validity: Then and now Retrieved February, 2012,

from http://www.albany.edu/cpr/brunswik/notes/essay2.html

Hanin, Y. L. (2000a). Successful and poor performance and emotions. In Y. L. Hanin

(Ed.), Emotions in Sport (pp. 157-187). Champaign, IL: Human Kinetics.

Hanin, Y. L. (2007a). Emotions and athletic performance: Individual zones of

optimal functioning. In D. Smith & M. Bar-Eli (Eds.), Essential Readings in

Sport and Exercise Psychology (pp. 55-74). Chamapign, IL: Human Kinetics.

Hanin, Y. L. (2007b). Emotions in Sport: Current Issues and Perspectives. In G.

Tenenbaum & R. C. Eklund (Eds.), Handbook of Sport Psychology (3rd ed.).

Hoboken: John Wiley & Sons, Inc.

Hanin, Y. L. (2010). Coping with anxiety in sport. In A. R. Nicholls (Ed.), Coping in

Sport: Theory, Methods, and Related Constructs (pp. 159-175). New York,

NY: Nova Science

http://www.albany.edu/cpr/brunswik/notes/essay2.html

Bibliography 243

Hanin, Y. L. (Ed.). (2000b). Emotions in Sport. Champaign, IL: Human Kinetics.

Hanin, Y. L., & Hanina, M. (2009). Optimization of performance in top-level

athletes: An action-focused coping approach. International Journal of Sports

Science & Coaching, 4, 47-91. doi: 10.1260/1747-9541.4.1.47

Hanin, Y. L., & Syrjä, P. (1995). Performance affect in soccer players: An

application of the IZOF model. International Journal of Sports Medicine,

16(4), 260-265. doi: 10.1055/s-2007-973002

Hardy, L. (1996). Testing the predictions of the Cusp Catastrophe Model of anxiety

and performance. The Sport Psychologist, 10, 140-156.

Hardy, L., & Fazey, J. A. (1987). The inverted-U hypothesis: A catastrophe for sport

psychology? Paper presented at the Annual Conference of the North

American Society for the Psychology of Sport and Physical Activity,

Vancouver, Canada.

Hardy, L., & Parfitt, G. (1991). A catastrophe model of anxiety and performance.

British Journal of Psychology, 82, 163-178.

Headrick, J., Davids, K., Renshaw, I., Araújo, D., Passos, P., & Fernandes, O.

(2012). Proximity-to-goal as a constraint on patterns of behaviour in attacker-

defender dyads in team games. Journal of Sports Sciences, 30(3), 247-253.

doi: 10.1080/02640414.2011.640706


dynamics of expertise acquisition in sport: The role of affective learning

design. Psychology of Sport and Exercise, 16, 83-90. doi:

10.1016/j.psychsport.2014.08.006

Headrick, J., Renshaw, I., Pinder, R. A., & Davids, K. (2012). Attunement to haptic

information helps skilled performers select implements for striking a ball in

cricket. Attention, Perception & Psychophysics, 74(8), 1782-1791. doi:

10.3758/s13414-012-0368-y

Heft, H. (2010). Affordances and the perception of landscape: An inquiry into

environmental perception and aesthetics. In C. Ward Thompson, P. Aspinall

& S. Bell (Eds.), Open Space: People Space 2, Innovative Approaches to

Researching Landscape and Health (pp. 9-32). Abingdon: Routledge.

Hilgard, E. R. (1980). The trilogy of the mind: Cognition, affection, and conation.

Journal of the History of the Behavioral Sciences, 16, 107-117.

Hollis, G., Kloos, H., & Van Orden, G. C. (2009). Origins of order in cognitive

activity. In S. J. Guastello, M. Koopmans & D. Pincus (Eds.), Chaos and

Complexity in Psychology: The Theory of Nonlinear Dynamical Systems (pp.

206-241). New York, NY: Cambridge University Press.

Hong, S. L., & Newell, K. M. (2006). Change in the organization of degrees of

freedom with learning. Journal of Motor Behavior, 38(2), 88-100.

Hristovski, R., Davids, K., & Araújo, D. (2009). Information for regulating action in

sport: Metastability and emergence of tactical solutions under ecological

constraints. In D. Araújo, H. Ripoll & M. Raab (Eds.), Perspectives on

Cognition and Action in Sport. New York: Nova Sciences Publishers Inc.

Hristovski, R., Davids, K., Araújo, D., & Button, C. (2006). How boxers decide to

punch a target: Emergent behaviour in nonlinear dynamical movement

systems. Journal of Sports Science and Medicine, 5, 60-73.

Hristovski, R., Davids, K., Araújo, D., & Passos, P. (2011). Constraints-induced

emergence of functional novelty in complex neurobiological systems: A basis

for creativity in sport. Nonlinear Dynamics, Psychology, and Life Sciences,

15(2), 175-206.

244 Bibliography

Hristovski, R., Davids, K., Passos, P., & Araújo, D. (2012). Sport performance as a

domain of creative problem solving for self-organizing performer-

environment systems. The Open Sports Sciences Journal, 5(Suppl 1), 26-35.

doi: 10.2174/1875399X01205010026

Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure

analysis: Conventional criteria versus new alternatives. Structural Equation

Modelling: A Multidisciplinary Journal, 6, 1-55. doi:

10.1080/10705519909540118

Hutto, D. D. (2012). Truly enactive emotion. Emotion Review, 4(2), 176-181. doi:

10.1177/1754073911430134

IBM Corp. (2013). IBM SPSS AMOS Statistics for Windows. (Version 22.0).

Armonk, NY: IBM Corp.

IBM Corp. (2013). IBM SPSS Statistics for Windows. (Version 22.0). Armonk, NY:

IBM Corp.

Isberg, L. (2000). Anger, aggressive behavior, and athletic performance. In Y. L.

Hanin (Ed.), Emotions in Sport (pp. 113-133). Champaign, IL: Human

Kinetics.

Jackson, D. L., Gillaspy, J. A., & Purc-Stephenson, R. (2009). Reporting practices in

confirmatory factor analysis: An overview and some recommendations.

Psychological Methods, 14, 6-23. doi: 10.1037/a0014694

Jackson, S. A. (2000). Joy, fun, and flow state in sport. In Y. L. Hanin (Ed.),

Emotionin in Sport (pp. 135-155). Champaign, IL: Human Kinetics.

Jackson, S. A., & Marsh, H. W. (1996). Development and validation of a scale to

measure optimal experience: The flow state scale. Journal of Sport &

Exercise Psychology, 18, 17-35.

James, W. (1884). What is an emotion? Mind, 9(34), 188-205.

Janelle, C. M., Singer, R. N., & Williams, A. M. (1999). External distraction and

attentional narrowing: Visual search evidence. Journal of Sport & Exercise

Psychology, 21, 70-91.

Jarvilehto, T. (1998a). The theory of the organism-environment system: I.

Description of the theory. Integrative Physiological and Behavioral Science,

33(4), 321-334.

Jarvilehto, T. (1998b). The theory of the organism-environment system: II.

Significance of nervous activity in the organism-environment system.

Integrative Physiological and Behavioral Science, 33(4), 335-342.

Jarvilehto, T. (1999). The theory of the organism-environment system: III. Role of

efferent influences on receptors in the formation of knowledge. Integrative

Physiological and Behavioral Science, 34(2), 90-100.

Jarvilehto, T. (2000a). Feeling as knowing - Part 1: Emotion as reorganization of the

organism-environment system. Consciousness and Emotion, 1, 53-65.

Jarvilehto, T. (2000b). Theory of the organism-environment system: IV. The

problem on mental activity and consciousness. Integrative Physiological and

Behavioral Science, 35(2), 35-57.

Jarvilehto, T. (2001). Feeling as knowing - Part 2: Emotion, consciousness, and brain

activity. Emotion and Consciousness, 2, 75-102.

Jarvilehto, T. (2009). The theory of the organism-environment system as a basis of

experimental work in psychology. Ecological Psychology, 21(2), 112-120.

doi: 10.1080/10407410902877066

Bibliography 245

Jones, G. (1995). More than just a game: Research developments and issues in

competitive anxiety in sport. British Journal of Psychology, 86, 449-478. doi:

10.1111/j.2044-8295.1995.tb02565.x

Jones, M. V. (2003). Controlling emotions in sport. The Sport Psychologist, 17, 471-

486.

Jones, M. V., Lane, A. M., Bray, S. R., Uphill, M. A., & Catlin, J. (2005).

Development and validation of the sport emotion questionnaire. Journal of

Sport & Exercise Psychology, 27(4), 407-431.

Juarrero, A. (2000). Dynamics in action: Intentional behaviour as a complex system.

Emergence, 2(2), 24-57.

Kelso, J. A. S. (1990). Phase transitions: Foundations of behaviour. In H. Haken

(Ed.), Synergetics of Cognition (pp. 249-268). Berlin, Germany: Springer.

Kelso, J. A. S. (1994). The informational character of self-organized coordination

dynamics. Human Movement Science, 13, 393-413.

Kelso, J. A. S. (1995). Dynamic Patterns: The Self-Organization of Brain and

Behavior. Cambridge: MIT press.

Kelso, J. A. S. (2003). Cognitive coordination dynamics. In W. Tshacher & J. P.

Dauwalder (Eds.), The Dynamical Systems Approach to Cognition: Concepts

and Empirical Paradigms Based on Self-Organisation, Embodiment and

Coordination Dynamics. (pp. 45-71). Singapore: World Scientific Publishing.

Kelso, J. A. S. (2008). An essay on understanding the mind. Ecological Psychology,

20(2), 180-208. doi: 10.1080/10407410801949297

Kelso, J. A. S. (2012). Multistability and metastability: understanding dynamic

coordination in the brain. Philosophical Transactions of the Royal Society of

London, Series B: Biological Sciences, 367, 906-918.

Kelso, J. A. S., & Tognoli, E. (2009). Toward a complementary neuroscience:

Metastable coordination dynamics of the brain. In N. Murphy, G. F. R. Ellis

& T. O'Connor (Eds.), Downward Causation and the Neurobiology of Free

Will (pp. 103-124). Berlin, Heidelberg: Springer.

Kelso, J. A. S., & Zanone, P. G. (2002). Coordination dynamics of learning and

transfer across different effector systems. Journal of Experimental

Psychology: Human Perception and Performance, 28(4), 776-797. doi:

10.1037//0096-1523.28.4.776

Keltner, D., & Gross, J. J. (1999). Functional accounts of emotions. Cognition &

Emotion, 13(5), 467-480. doi: 10.1080/026999399379140

Kirlik, A. (2009). Brunswikian resources for event-perception research. Perception,

38, 376-398.

Kiverstein, J., & Miller, M. (2015). The embodied brain: Towards a radical

embodied cognitive neuroscience. Frontiers in Human Neuroscience, 9,

article 237. doi: 10.3389/fnhum.2015.00237

Komar, J., Chow, J. Y., Chollet, D., & Seifert, L. (2015). Neurobiological

degeneracy: Supporting stability, flexibility and pluripotentiality in complex

motor skill. Acta Psychologica, 154, 26-35. doi:

10.1016/j.actpsy.2014.11.002

LaBar, K. S., & Cabeza, R. (2006). Cognitive neuroscience of emotional memory.

Nature Reviews Neuroscience, 7, 54-64. doi: 10.1038/nrn1825

Lane, A. M., Beedie, C. J., Jones, M. V., Uphill, M. A., & Devenport, T. J. (2012).

The BASES expert statement on emotion regulation in sport. Journal of

Sports Sciences, 30(11), 1189-1195. doi: 10.1080/02640414.2012.693621

246 Bibliography

Lane, A. M., & Terry, P. C. (2000). The nature of mood: Development of a

conceptual model with a focus on depression. Journal of Applied Sport

Psychology, 12, 16-33. doi: 10.1080/10413200008404211

Larsen, R. J., & Diener, E. (1987). Affect intensity as an individual difference

characteristic: A review. Journal of Research in Personality, 21, 1-39. doi:

10.1016/0092-6566(87)90023-7

Larsen, R. J., Diener, E., & Cropanzano, R. S. (1987). Cognitive operations

associated with individual differences in affect intensity. Journal of Social

Psychology, 53(4), 767-774. doi: 10.1037/0022-3514.53.4.767

Lazarus, R. S. (1991). Emotion and Adaptation. New York, NY: Oxford University

Press.

Lazarus, R. S. (2000). How emotions influence performance in competitive sports.

The Sport Psychologist, 14, 229-252.

LeDoux, J. E. (2000). Emotion circuits in the brain. Annual Review of Neuroscience,

23, 155-184.

Lepper, M. R. (1994). "Hot" versus "Cold" cognition: An Abelsonian voyage. In R.

C. Schank & E. Langer (Eds.), Beliefs, Reasoning, and Decision making:

Psycho-Logic in Honor of Bob Abelson (pp. 237-273). Hillsdale, NJ:

Lawrence Erlbaum Associates.

Lewis, M. D. (1996). Self-organising cognitive appraisals. Cognition and Emotion,

10, 1-26. doi: 10.1080/026999396380367

Lewis, M. D. (1997). Personality self-organization: Cascading constraints on

cognition-emotion interaction. In A. Fogel, M. C. Lyra & J. Valsiner (Eds.),

Dynamics and Indeterminism in Developmental and Social Processes (pp.

193-216). Mahwah, NJ: Erlbaum.

Lewis, M. D. (2000a). Emotional self-organization at three time scales. In M. D.

Lewis & I. Granic (Eds.), Emotion, Development, and Self-Organization:

Dynamic Systems Approaches to Emotional Development (pp. 37-69).


Lewis, M. D. (2000b). The promise of dynamic systems approaches for an integrated

account of human development. Child Development, 71, 36-43. doi:

10.1111/1467-8624.00116

Lewis, M. D. (2002). Interacting time scales in personality (and cognitive)

development: Intentions, emotions, and emergent forms. In N. Granott & J.

Parziale (Eds.), Microdevelopment: Transition Processes in Development and

Learning (pp. 183-212). Cambridge: Cambridge University Press.

Lewis, M. D. (2004). The emergence of mind in the emotional brain. In A.

Demetriou & A. Raftopoulos (Eds.), Cognitive developmental change. New

York Cambridge University Press.

Lewis, M. D. (2005). Bridging emotion theory and neurobiology through dynamic

systems modeling. Behavioral and Brain Sciences, 28(2), 169-194. doi:

10.1017/S0140525X0500004X

Lewis, M. D., & Cook, M. L. (2007). Changing habits of emotion regulation at

transition points in infancy: A dynamic systems analysis. Journal of

Developmental Processes, 3, 67-89.

Lewis, M. D., & Granic, I. (2000). Introduction: A new approach to the study of

emotional development. In M. D. Lewis & I. Granic (Eds.), Emotion,

Development, and Self-Organization: Dynamic Systems Approaches to

Emotional Development (pp. 1-12). Cambridge, UK: Cambridge University

Press.

Bibliography 247

Lewis, M. D., & Stieben, J. (2004). Emotion regulation in the brain: Conceptual

issues and direction for developmental research. Child Development, 75(2),

371-376. doi: 10.1111/j.1467-8624.2004.00680.x

Lewis, M. D., & Todd, R. M. (2005). Getting emotional: A neural perspective on

emotion, intention and consciousness. Journal of Consciousness Studies,

12(8-10), 210-235.

Liu, Y.-T., Mayer-Kress, G., & Newell, K. M. (2006). Qualitative and quantitative

change in the dynamics of motor learning. Journal of Experimental

Psychology, 32(2), 380-393. doi: 10.1037/0096-1523.32.2.380

Mace, R., Eastman, C., & Carroll, D. (1986). Stress inoculation training: A case

study in gymnastics. British Journal of Sports Medicine, 20(3), 139-141. doi:

10.1136/bjsm.20.3.139

Mann, D. L., Abernethy, B., & Farrow, D. (2010). The resilience of natural

interceptive actions to refractive blur. Human Movement Science, 29(3), 386-

400. doi: 10.1016/j.humov.2010.02.007

Martens, R., Burton, D., Vealey, R. S., Bump, L. A., & Smith, D. E. (1990).

Development and validation of the Competitive State Anxiety Inventory-2. In

R. Martens, R. S. Vealey & D. Burton (Eds.), Competitive Anxiety in Sport

(pp. 117-190). Champaign, Il: Human Kinetics.

Martens, R., Vealey, R. S., & Burton, D. (1990). Competitive Anxiety in Sport.

Champaign, IL: Human Kinetics.

Masters, R. S. W., Poolton, J. M., Maxwell, J. P., & Raab, M. (2008). Implicit motor

learning and complex decision making in time-constrained environments.

Journal of Motor Behavior, 40, 71-79. doi: 10.3200/JMBR.40.1.71-80

Mathews, A., & MacLeod, C. (1994). Cognitive approaches to emotion and

emotional disorders. Annual Review of Psychology, 45, 25-50.

McCarthy, P. J. (2011). Positive emotion in sport performance: current status and

future directions. International Review of Sport and Exercise Psychology, 4,

50-69. doi: 10.1080/1750984X.2011.560955

McClelland, D. C., Atkinson, J. W., Clark, R. A., & Lowell, E. L. (1953). The

achievement motive. New York: Appleton-Century-Crofts.

McNair, D. M., Lorr, M., & Droppleman, L. F. (1971). Profile of Mood State

Manual. San Diego, CA: Educational and Industrial Testing Service.

McPherson, S. L. (2008). Tactics: Using knowledge to enhance performance. In D.

Farrow, J. Baker & C. MacMahon (Eds.), Developing Elite Sports

Performers: Lessons From Theory and Practice (pp. 155-171). London:

Routledge.

McPherson, S. L., & MacMahon, C. (2008). How baseball players prepare to bat:

Tactical knowledge as a mediator of expert performance in baseball. Journal

of Sport & Exercise Psychology, 30(6), 755-778.

Millar, S. K., Oldham, A. R. H., & Donovan, M. (2011). Coaches' self-awareness of

timing, nature and intent of verbal instructions to athletes. International

Journal of Sports Science & Coaching, 6(4), 503-513. doi: 10.1260/1747-

9541.6.4.503

Moll, T., Jordet, G., & Pepping, G. J. (2010). Emotional contagion in soccer penalty

shootouts: Celebration of individual success is associated with ultimate team

success. Journal of Sports Sciences, 28(9), 983-992. doi:

10.1080/02640414.2010.484068

Moy, B., Renshaw, I., & Davids, K. (2014). Variations in acculturation and

Australian physical education teacher education students' receptiveness to an

248 Bibliography

alternative pedagogical approach to games teaching. Physical Education and

Sport Pedagogy, 19(4), 349-369. doi: 10.1080/17408989.2013.780591

Moy, B., Renshaw, I., & Davids, K. (in press). The impact of nonlinear pedagogy on

physical education teacher education student's intrinsic motivation. Physical

Education and Sport Pedagogy.

Müller, S., & Abernethy, B. (2006). Batting with occluded vision: An in situ

examination of the information pick-up and interceptive skills of high- and

low-skilled cricket batsmen. Journal of Science and Medicine in Sport, 9(6),

446-458.

Müller, S., & Abernethy, B. (2008). Validity and reliability of a simple categorical

tool for the assessment of interceptive skill. Journal of Science and Medicine

in Sport, 11(6), 549-552. doi: 10.1016/j.jsams.2007.08.003

Müller, S., & Abernethy, B. (2012). Expert anticipatory skill in striking sports: A

review and a model. Research Quarterly for Exercise and Sport, 83(2), 175-

187.

Nesse, R. M., & Ellsworth, P. C. (2009). Evolution, emotions, and emotional

disorders. American Psychologist, 64(2), 129-139. doi: 10.1037/a0013503

Newell, K. M. (1986). Constraints on the development of coordination. In M. G.

Wade & H. T. A. Whiting (Eds.), Motor development in children: Aspects of

coordination and control (pp. 341-360). Dordrecht: Martinus Nijhoff.

Newell, K. M., Broderick, M. P., Deutsch, K. M., & Slifkin, A. B. (2003). Task goals

and change in dynamical degrees of freedom with motor learning. Journal of

Experimental Psychology: Human Perception and Performance, 29(2), 379-

387.

Newell, K. M., Liu, Y.-T., & Mayer-Kress, G. (2001). Time scales in motor learning

and development. Psychological Review, 108, 57-82.

Newell, K. M., & Ranganathan, R. (2010). Instructions as constraints in motor skill

acquisition. In I. Renshaw, K. Davids & G. J. P. Savelsbergh (Eds.), Motor

Learning in Practice: A Constraints-led Approach (pp. 17-32). New York,

NY: Routledge.

Newell, K. M., & Valvano, J. (1998). Therapeutic intervention as a constraint in

learning and relearning movement skills. Scandinavian Journal of

Occupational Therapy, 5, 51-57.

Nibbeling, N., Daanen, H. A. M., Gerritsma, R. M., Hofland, R. M., & Oudejans, R.

R. D. (2012). Effects of anxiety on running with and without an aiming task.

Journal of Sports Sciences, 30, 11-19. doi: 10.1080/02640414.2011.617386

Nibbeling, N., Oudejans, R. R. D., & Daanen, H. A. M. (2012). Effects of anxiety, a

cognitive secondary task, and expertise on gaze behaviour and performance

in a far aiming task. Psychology of Sport and Exercise, 13, 427-435. doi:

10.1016/j.psychsport.2012.02.002

Nicholls, J. G. (1984). Achievement motivation: Conceptions of ability, subjective

experience, task choice, and performance. Psychological Review, 91(3), 328-

346.

Niedenthal, P. M. (2008). Emotion concepts. In M. Lewis, J. M. Haviland - Jones &

L. Feldman Barrett (Eds.), Handbook of Emotions (pp. 587-600). New York,

NY: Guilford Press.

Nieuwenhuys, A., & Oudejans, R. R. D. (2011). Training with anxiety: Short- and

long-term effects on police officers' shooting behaviour under pressure.

Cognitive Processing, 12(3), 277-288. doi: 10.1007/s10339-011-0396-x

Bibliography 249

Oatley, K., & Jenkins, J. M. (1992). Human emotions: Function and dysfunction.

Annual Review of Psychology, 43(55-85).

Oatley, K., & Jenkins, J. M. (1996). Understanding emotions. Cambridge, MA:

Blackwell Scientific.

Ochsner, K. N., & Gross, J. J. (2007). The neural architecture of emotion regulation.

In J. J. Gross (Ed.), Handbook of emotion regulation. London: Guilford Press.

Öhman, A. (2008). Fear and anxiety: Overlaps and dissociations. In M. Lewis, J. M.

Haviland - Jones & L. Feldman Barrett (Eds.), Handbook of Emotions (pp.

709-729). New York, NY: Guilford Press.

Öhman, A., Flykt, A., & Esteves, F. (2001). Emotion drives attention: Detecting the

snake in the grass. Journal of Experimental Psychology: General, 130(3),

466-478. doi: 10.1037/AXJ96-3445.130.3.466

Öhman, A., & Mineka, S. (2001). Fears, phobias, and preparedness: Toward and

evolved module of fear and fear learning. Psychological Review, 108(3), 483-

522. doi: 10.1037//0033-295X.108.3.483

Orth, D., Davids, K., Araújo, D., Renshaw, I., & Passos, P. (2014). Effects of a

defender on run-up velocity and ball speed when crossing a football.

European Journal of Sport Science, 14(sup1), S316-S323. doi:

10.1080/17461391.2012.696712

Ortony, A., Clore, G. L., & Collins, A. (1994). The Cognitive Structure of Emotions.

Cambridge: Cambridge University Press.

Oudejans, R. R. D. (2008). Reality-based practice under pressure improves handgun

shooting performanc of police officers. Ergonomics, 51(3), 261-273. doi:

10.1080/00140130701577435

Oudejans, R. R. D., Michaels, C. F., & Bakker, F. C. (1997). The effects of baseball

experience on movement initiation in catching fly balls. Journal of Sports

Sciences, 15(6), 587-595.

Oudejans, R. R. D., & Nieuwenhuys, A. (2009). Perceiving and moving in sports and

other high-pressure contexts. Progress in Brain Research, 174, 35-48. doi:

10.1016/S0079-6123(09)01304-1

Oudejans, R. R. D., & Pijpers, J. R. (2009). Training with anxiety has a positive

effect on expert perceptual-motor performance under pressure. The Quarterly

Journal of Experimental Psychology, 62(8), 1631-1647. doi:

10.1080/17470210802557702

Oudejans, R. R. D., & Pijpers, J. R. (2010). Training with mild anxiety may prevent

choking under higher levels of anxiety. Psychology of Sport and Exercise, 11,

44-50. doi: 10.1016/j.psychsport.2009.05.002

Padmala, S., & Pessoa, L. (2008). Affective learning enhances visual detection and

responses in primary visual cortex. The Journal of Neuroscience, 28(24),

6202-6210. doi: 10.1523/JNEUROSCI.1233-08.2008

Panksepp, J. (1998). Affective Neuroscience. Oxford: Oxford University Press.

Parrott, W. G. (2001). The Nature of Emotion. In A. Tesser & N. Schwarz (Eds.),

Blackwell Handbook of Social Psychology: Intraindividual Processes (pp.

375-390). Malden, MA: Blackwell Publishers.

Passos, P., Araújo, D., & Davids, K. (2013). Self-organization processes in field-

invasion team sports. Sports Medicine, 43, 1-7. doi: 10.1007/s40279-012-

0001-1

Passos, P., Araújo, D., Davids, K., Gouveia, L., Milho, J., & Serpa, S. (2008).

Information-governing dynamics of attacker-defender interactions in youth

rugby union. Journal of Sports Sciences, 26(13), 1421-1429.

250 Bibliography

Pessoa, L. (2008). On the relationship between emotion and cognition. Nature

Reviews Neuroscience, 9, 148-158.

Pessoa, L. (2011). Reprint of: Emotion and cognition and the amygdala: From "what

is it?" to "what's to be done?" Neuropsychologia, 49(4), 681-694.

Phillips, E., Davids, K., Araújo, D., & Renshaw, I. (2014). Talent Development and

Expertise in Sport. In K. Davids, R. Hristovski, D. Araújo, N. Balagué Serre,

C. Button & P. Passos (Eds.), Complex Systems in Sport (pp. 241-260).

London: Routledge.

Phillips, E., Davids, K., Renshaw, I., & Portus, M. (2010). Expert performance in

sport and the dynamics of talent development. Sports Medicine, 40(4), 271-

283.

Pijpers, J. R., Oudejans, R. R. D., & Bakker, F. C. (2005). Anxiety-induced changes

in movement behaviour during the execution of a complex whole-body task.

The Quarterly Journal of Experimental Psychology, 58A(3), 421-445. doi:

10.1080/02724980343000945

Pijpers, J. R., Oudejans, R. R. D., Bakker, F. C., & Beek, P. J. (2006). The role of

anxiety in perceiving and realizing affordances. Ecological Psychology,

18(3), 131-161. doi: 10.1207/s15326969eco1803_1

Pijpers, J. R., Oudejans, R. R. D., Holsheimer, F., & Bakker, F. C. (2003). Anxiety-

performance relationships in climbing: A process-orientated approach.

Psychology of Sport and Exercise, 4(3), 283-304. doi: 10.1016/S1469-

0292(02)00010-9

Pinder, R. A. (2012). Representative Learning Design in Dynamic Interceptive

Actions. Doctor of Philosophy, Queensland University of Technology,

Brisbane.

Pinder, R. A., Davids, K., & Renshaw, I. (2012). Metastability and emergent

performance of dynamic interceptive actions. Journal of Science and

Medicine in Sport, 15(5), 437-443. doi: 10.1016/j.jsams.2012.01.002

Pinder, R. A., Davids, K., Renshaw, I., & Araújo, D. (2011a). Manipulating

informational constraints shapes movement reorganization in interceptive

actions. Attention, Perception & Psychophysics, 73, 1242 - 1254. doi:

10.3758/s13414-011-0102-1

Pinder, R. A., Davids, K., Renshaw, I., & Araújo, D. (2011b). Representative

learning design and functionality of research and practice in sport. Journal of

Sport & Exercise Psychology, 33, 146-155.


developing representative tasks in sport. In D. Farrow & J. Baker (Eds.),


Pinder, R. A., Renshaw, I., & Davids, K. (2009). Information-movement coupling in

developing cricketers under changing ecological practice constraints. Human

Movement Science, 28, 468-479. doi: 10.1016/j.humov.2009.02.003

Pinder, R. A., Renshaw, I., & Davids, K. (2013). The role of representative design in

talent development: a comment on “Talent identification and promotion

programmes of Olympic athletes”. Journal of Sports Sciences, 31(8), 803-

806. doi: 10.1080/02640414.2012.718090

Pinder, R. A., Renshaw, I., Davids, K., & Kerhervé, H. (2011). Principles for the use

of ball projection machines in elite and developmental sport programmes.

Sports Medicine, 41(10), 793-800.



Bibliography 251

Balagué Serre, C. Button & P. Passos (Eds.), Complex Systems in Sport (pp.


Portus, M. R., & Farrow, D. (2011). Enhancing cricket batting skill: implications for

biomechanics and skill acquisition research and practice. Sports

Biomechanics, 10(4), 294-305.

Proffitt, D. R., Bhalla, M., Gossweiler, R., & Midgett, J. (1995). Perceiving

geographical slant. Psychonomic Bulletin & Review, 2(4), 409-428. doi:

10.3758/BF03210980

Raedeke, T. D. (2007). The relationship between enjoyment and affective responses

to exercise. Journal of Applied Sport Psychology, 19, 105-115. doi:

10.1080/10413200601113638

Raglin, J. S., & Hanin, Y. L. (2000). Competitive anxiety. In Y. L. Hanin (Ed.),

Emotions in Sport (pp. 93-112). Champaign, IL: Human Kinetics.

Renshaw, I., Chappell, G. S., Fitzgerald, D., & Davison, J. (2010). The battle zone:

constraint-led coaching in action. Paper presented at the Conference of

Science, Medicine & Coaching in Cricket, Gold Coast, Queensland:

Australia.

Renshaw, I., Chow, J. Y., Davids, K., & Hammond, J. (2010). A constraints-led

perspective to understanding skill acquisition and game play: A basis for

integration of motor learning theory and physical education praxis? Physical

Education & Sport Pedagogy, 15(2), 117-137.

Renshaw, I., & Davids, K. (2004). Nested task constraints shape continuous

perception-action coupling control during human locomotor pointing.

Neuroscience Letters, 369, 93-98.

Renshaw, I., & Davids, K. (2014). Task Constraints. In R. C. Eklund & G.

Tenenbaum (Eds.), Encyclopaedia of Sport and Exercise Psychology (Vol. 2,

pp. 734-737). NY: Sage Publications.

Renshaw, I., Davids, K., Shuttleworth, R., & Chow, J. Y. (2009). Insights from

ecological psychology and dynamical systems theory can underpin a

philosophy of coaching. International Journal of Psychology, 40(4), 540-602.

Renshaw, I., & Fairweather, M. M. (2000). Cricket bowling deliveries and the

discrimination ability of professional and amateur batters. Journal of Sports

Sciences, 18(12), 951-957.




France.

Renshaw, I., Oldham, A. R. H., & Bawden, M. (2012). Nonlinear pedagogy

underpins intrinsic motivation in sports coaching. The Open Sports Sciences

Journal, 5(Suppl 1), 88-99. doi: 10.2174/1875399X01205010088

Renshaw, I., Oldham, A. R. H., Davids, K., & Golds, T. (2007). Changing ecological

constraints of practice alters coordination of dynamics interceptive actions.

European Journal of Sport Science, 7(3), 157-167. doi:

10.1080/17461390701643026

Renshaw, I., Oldham, A. R. H., Glazier, P., & Davids, K. (2004). Around the world:

Why applied sport scientists need a theoretical model of the performer. The

Sport and Exercise Scientist, 1, 24.

Riley, M. A., Shockley, K., & Van Orden, G. (2012). Learning from the body about

the mind. Topics in Cognitive Science, 4, 21-34. doi: 10.1111/j.1756-

8765.2011.01163.x

252 Bibliography

Roe, J., & Aspinall, P. (2011). The emotional affordances of forest settings: An

investigation in boys with extreme behavioural problems. Landscape

Research, 36(5), 535-552.

Ryan, R. M., & Deci, E. L. (2000a). Intrinsic and extrinsic motivations: Classic

definitions and new directions. Contemporary Educational Psychology, 25,

54-67.

Ryan, R. M., & Deci, E. L. (2000b). Self-determination theory and the facilitation of

intrinsic motivation, social development, and well-being. American

Psychologist, 55, 68-78.

Sagar, S. S., Lavallee, D., & Spray, C. M. (2007). Why young athletes fear failure:

Consequences of failure. Journal of Sports Sciences, 25(11), 1171-1184. doi:

10.1080/02640410601040093

Schmidt, R. A., & Wrisberg, C. A. (2008). Motor Learning and Performance: A

Situation-Based Learning Approach (4th ed.). Champaign, IL: Human

Kinetics.

Schöner, G., Zanone, P. G., & Kelso, J. A. S. (1992). Learning as change of

coordination dynamics: Theory and experiment. Journal of Motor Behavior,

24, 29-48. doi: 10.1080/00222895.1992.9941599

Schreiber, J. B., Nora, A., Stage, F. K., Barlow, E. A., & King, J. (2006). Reporting

structural equation modeling and confirmatory factor analysis results: A

review. The Journal of Educational Research, 99(6), 323-338. doi:

10.3200/JOER.99.6.323-338

Schwarz, N. (2000). Emotion, cognition, and decision making. Cognition and

Emotion, 14(4), 433-440.

Searle, J. R. (1983). Intentionality: An essay in the philosophy of mind. Cambridge,

UK: Cambridge University Press.

Seifert, L., Button, C., & Davids, K. (2013). Key properties of expert movement

systems in sport: An ecological dynamics perspective. Sports Medicine,

43(3), 167-178. doi: 10.1007/s40279-012-0011-z

Seifert, L., & Davids, K. (2012). Intentions, perceptions and actions constrain

functional intra - and inter - individual variability in the acquisition of

expertise in individual sports. The Open Sports Sciences Journal, 5(Suppl 1),

68-75. doi: 10.2174/1875399X01205010068

Seifert, L., Wattlebled, L., L'Hermette, M., Bideault, G., Herault, R., & Davids, K.

(2013). Skill transfer, affordances and dexterity in different climbing

environments. Human Movement Science, 32(6), 1339-1352. doi:

10.1016/j.humov.2013.06.006

Shaver, P., Schwartz, J., Kirson, D., & O'Connor, C. (1987). Emotion knowledge:

Further exploration of a prototype approach. Journal of Personality and

Social Psychology, 52(6), 1061-1086. doi: 10.1037/0022-3514.52.6.1061

Shaw, R. (2001). Processes, acts, and experiences: Three stances on the problem of

intentionality. Ecological Psychology, 13(4), 275-314. doi:

10.1207/S15326969ECO1304_02

Shaw, R., & Turvey, M. T. (1999). Ecological foundations of cognition: II. Degree of

freedom and conserved quantities in animal-environment systems. Journal of

Consciousness Studies, 6(11-12), 111-124.

Simon, H. A. (1956). Rational choice and the structure of the environment.

Psychological Review, 63(2), 129-138. doi: 10.1037/h0042769

Simon, H. A. (1990). Invariants of human behaviour. Annual Review of Psychology,

41, 1-19. doi: 10.1146/annurev.ps.41.020190.000245

Bibliography 253

Skinner, N., & Brewer, N. (2004). Adaptive approaches to competition: Challenge

appraisals and positive emotion. Journal of Sport & Exercise Psychology, 26,

283-305.

Smith, L. B., & Thelen, E. (2003). Development as a dynamic system Trends in

Cognitive Sciences, 7(8), 343-348. doi: 10.1016/S1364-6613(03)00156-6

Smith, R. E., Smoll, F. L., & Schutz, R. W. (1990). Measurement and correlates of

sport-specific cognitive and somatic trait anxiety: The sport anxiety scale.

Anxiety Research, 2(4), 263-280. doi: 10.1080/08917779008248733

Solomon, R. C. (2008). The philosophy of emotions. In M. Lewis, J. M. Haviland -

Jones & L. Feldman Barrett (Eds.), Handbook of Emotions (3rd ed., pp. 3-

16). New York: The Guilford Press.

Spielberger, C. D. (1966). Anxiety and Behavior. New York, NY: Academic Press.

Spinoza, B. (1674/2006). Part III: Concerning the Origin and Nature of the Emotions

(S. Shirley, Trans.). In M. L. Morgan (Ed.), The Essential Spinoza: Ethics

and Related Writings (pp. 61-101). Indianapolis, IN: Hackett Publishing

Company, Inc.

Steiger, J. H. (1990). Structural model evaluation and modification: An interval

estimation approach. Multivariate Behavioral Research, 25(2), 173-180. doi:

10.1207/s15327906mbr2502_4

Steiger, J. H., & Lind, J. C. (1980). Statistically based tests for the number of

common factors. Paper presented at the Pychometric Society annual meeting,

Iowa City, IA.

Stoffregen, T. A., Bardy, B. G., Smart, L. J., & Pagulayan, R. (2003). On the nature

and evaluation of fidelity in virtual environments. In L. J. Hettinger & M. W.

Haas (Eds.), Virtual and Adaptive Environments: Applications, Implications

and Human Performance Issues (pp. 111-128). Mahwah: Lawrence Erlbaum

Associates.

Stone, J. A., Maynard, I., North, J. S., Panchuk, D., & Davids, K. (2015).

(De)synchronization of advanced visual information and ball flight

characteristics constrains emergent information–movement couplings during

one‑handed catching. Experimental Brain Research, 233, 449-458. doi:

10.1007/s00221-014-4126-3

Stretch, R., Bartlett, R., & Davids, K. (2000). A review of batting in men's cricket.

Journal of Sports Sciences, 18(12), 931 - 949. doi:

10.1080/026404100446748

Stretch, R., Buys, F., Du Toit, E., & Viljoen, G. (1998). Kinematics and kinetics of

the drive off the front foot in cricket batting. Journal of Sports Sciences, 16,

711-720. doi: 10.1080/026404198366344

Syrjä, P., & Hanin, Y. L. (1998). Individualized and group-orientated measures of

emotion in sport: A comparative study. Journal of Sports Sciences, 16(5),

398-399. doi: 10.1080/026404198366542

Tenenbaum, G., Basevitch, I., Gershgoren, L., & Filho, E. (2013). Emotions-

decision-making in sport: Theoretical conceptualization and experimental

evidence. International Journal of Sport and Exercise Psychology, 11(2),

151-168. doi: 10.1080/1612197X.2013.773687

Tenenbaum, G., & Elran, E. (2003). Congruence between actual and retrospective

reports of emotions for pre- and postcompetition states. Journal of Sport &

Exercise Psychology, 25(3), 323-340.

Thelen, E. (1995). Motor development: A new synthesis. American Psychologist,

50(2), 79-95. doi: 10.1037/0003-066X.50.2.79

254 Bibliography

Thelen, E. (2002). Self-organisation in developmental processes: Can systems

approaches work? In M. H. Johnson, Y. Munakata & R. O. Gilmore (Eds.),

Brain Development and Cognition: A reader (2nd ed., pp. 336-374). Malden,

MA: Blackwell.

Thelen, E., & Smith, L. B. (1994). A Dynamic Systems Approach to the Development

of Cognition and Action. Cambridge, MA: The MIT Press.

Todd, R. M., Talmi, D., Schmitz, T. W., Susskind, J., & Anderson, A. K. (2012).

Psychophysical and neural evidence for emotion-enhanced perceptual

vividness. The Journal of Neuroscience, 32(33), 11201-11212. doi: 10.1523/

JNEUROSCI.0155-12.2012

Toering, T., Elferink-Gemser, M. T., Jordet, G., Jorna, C., Pepping, G. J., &

Visscher, C. (2011). Self-regulation of practice behaviour among elite youth

soccer players: An exploratory observation study. Journal of Applied Sport

Psychology, 23, 110-128. doi: 10.1080/10413200.2010.534544

Travassos, B., Duarte, R., Vilar, L., Davids, K., & Araújo, D. (2012). Practice task

design in team sports: Representativeness enhanced by increasing

opportunities for action. Journal of Sports Sciences, 30(13), 1447-1454. doi:

10.1080/02640414.2012.712716

Tucker, D. M., Derryberry, D., & Luu, P. (2000). Anatomy and physiology of human

emotion: Vertical Integration of brainstem, limbic, and cortical systems. In J.

Borod (Ed.), Handbook of the Neuropsychology of Emotion. New York:

Oxford University Press.

Turvey, M. T. (2009). On the notion and implications of organism-environment

system. Ecological Psychology, 21(2), 97-111. doi:

10.1080/10407410902877041

Uphill, M. A., McCarthy, P. J., & Jones, M. V. (2009). Getting a grip on emotion

regulation in sport. In S. D. Mellalieu & S. Hanton (Eds.), Advances in

Applied Sport Psychology: A Review (pp. 163-194). London: Routledge.

Vallacher, R. R., & Nowak, A. (2009). The dynamics of human experience:

Fundamentals of dynamical social psychology. In S. J. Guastello, M.

Koopmans & D. Pincus (Eds.), Chaos and Complexity in Psychology: The

Theory of Nonlinear Dynamical Systems (pp. 370-401). New York, NY:

Cambridge University Press.

Vallerand, R. J., & Blanchard, C. M. (2000). The study of emotion in sport and

exercise: Historical, definitional, and conceptual perspectives. In Y. L. Hanin

(Ed.), Emotions in Sport (pp. 3-37). Champaign, IL: Human Kinetics.

van de Pol, P. K. C., & Kavussanu, M. (2011). Achievement goals and motivational

responses in tennis: Does the context matter? Psychology of Sport and

Exercise, 12(2), 176-183. doi: 10.1016/j.psychsport.2010.09.005

van de Pol, P. K. C., Kavussanu, M., & Ring, C. (2012a). The effects of training and

competition on achievement goals, motivational responses, and performance

in a golf-putting task. Journal of Sport & Exercise Psychology, 34(6), 787-

807.

van de Pol, P. K. C., Kavussanu, M., & Ring, C. (2012b). Goal orientations,

perceived motivational climate, and motivational outcomes in football: A

comparison between training and competition contexts. Psychology of Sport

and Exercise, 13, 491-499.

Van der Mars, H. (1989). Observer reliability: Issues and procedures. In P. W. Darst,

D. B. Zakrajsek & V. H. Mancini (Eds.), Analyzing Physical Education and

Sport Instruction (pp. 53-81). Champaign, IL: Human Kinetics.

Bibliography 255

Vickery, W., Dascombe, B., Duffield, R., Kellet, A., & Portus, M. (2013).

Battlezone: An examination of the physiological responses, movement

demands and reproducibility of small-sided cricket games. Journal of Sports

Sciences, 31, 77-86. doi: 10.1080/02640414.2012.720706

Vilar, L., Araújo, D., Davids, K., & Button, C. (2012). The role of ecological

dynamics in analysing performance in team sports. Sports Medicine, 42, 1-10.

doi: 10.2165/11596520

von Cranach, M., & Kalbermatten, U. (1982). Ordinary interactive action: Theory,

methods and some empirical findings. In M. von Cranach & R. Harré (Eds.),

The Analysis of action: Recent Theoretical and Emprical Advances (pp. 115-

160). Cambridge: Cambridge University Press.

Ward, J. H. (1963). Hierarchical grouping to optiimze an objective function. Journal

of the American Statistical Association, 58(301), 236-244. doi:

10.2307/2282967

Warr, P., Barter, J., & Brownbridge, G. (1983). On the independance of positive and

negative affect. Journal of Personality and Social Psychology, 44(3), 644-

651. doi: 10.1037/0022-3514.44.3.644

Warren, W. H. (2006). The dynamics of perception and action. Psychological

Review, 113(2), 358-389.

Watson, D., & Clark, L. A. (1994). Emotions, moods, traits, and temperaments:

Conceptual distinctions and empirical findings. In P. Ekman & R. J.

Davidson (Eds.), The Nature of Emotion: Fundamental Questions (pp. 89-

93). New York, NY: Oxford University Press.

Watson, D., & Clark, L. A. (1997). Measurement and mismeasurement of mood:

Recurrent and emergent issues. Journal of Personality Assessment, 68(2),

267-296. doi: 10.1207/s15327752jpa6802_4

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of

brief measures of positive and negative affect: The PANAS scales. Journal of

Personality and Social Psychology, 54(6), 1063-1070.

Weinberg, R. S., & Gould, D. (2015). Foundations of sport and exercise psychology

(6th ed.). Champaign, IL: Human Kinetics.

Williams, B., Onsman, A., & Brown, T. (2010). Exploratory factor analysis: A five-

step guide for novices. Journal of Emergency Primary Health Care, 8(3), 1-

13.

Withagen, R., de Poel, H. J., Araújo, D., & Pepping, G. J. (2012). Affordances can

invite behaviour: Reconsidering the relationship between affordances and

agency. New Ideas in Psychology, 30, 250-258.

Woodman, T., Davis, P. A., Hardy, L., Callow, N., Glasscock, I., & Yuill-Proctor, J.

(2009). Emotions and sport performance: An exploration of happiness, hope,

and anger. Journal of Sport & Exercise Psychology, 31, 169-188.

Woolmer, B., Noakes, T., & Moffett, H. (2008). Bob Woolmer's art and science of

cricket. London: New Holland.

Zadra, J. R., & Clore, G. L. (2011). Emotion and perception: The role of affective

information. Wiley Interdisciplinary Reviews: Cognitive Science, 2(6), 676-

685. doi: 10.1002/wcs.147

Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences.

American Psychologist, 35(2), 151-175. doi: 10.1037/0003-066X.35.2.151

Zanone, P. G., & Kelso, J. A. S. (1992). Evolution of behavioural attractors with

learning: Nonequilibrium phase transitions. Journal of Experimental

Psychology, 18(2), 403-421. doi: 10.1037/0096-1523.18.2.403

256 Bibliography

Appendices 257

Appendices

Appendix A – Selected interpretations of emotion

The following quotations represent a selection of interpretations, perspectives,

and opinions regarding the concept of emotion that reflect ideas and arguments

presented throughout this thesis supporting the interaction between emotions,

cognitions, and actions. Originating from a range of backgrounds and approaches,

spanning over three centuries, these perspectives and opinions exemplify the

difficulty of formulating a comprehensive and agreed definition of emotion.

“By emotion [affectus] I understand the affections of the body by

which the body’s power of activity is increased or diminished, assisted

or checked, together with the ideas of these affections. Thus, if we

can be the adequate cause of one of these affections, then by emotion I

understand activity, otherwise passivity.”

(Spinoza, 1674/2006, p. 62)

“My theory is that the bodily changes follow directly the perception of

the exciting fact and that our feeling of the same changes as they

occur is the emotion.”

(James, 1884, p. 204)

“Emotional phenomena are noninstrumental behaviours and

noninstrumental features of behaviour, physiological changes, and

evaluative, subject related experiences, as evoked be external or

258 Appendices

mental events, and primarily by the significance of such events. An

emotion is either an occurrence of phenomena of these three kinds or

the inner determinant of such phenomena; the choice will be made

later.”

(Frijda, 1986, p. 4)

“Emotion: Differently described and explained by different

psychologists, but all agree that it is a complex state of the organism,

involving bodily changes of a widespread character— in breathing,

pulse, gland secretion, etc. —and, on the mental side, a state of

excitement or perturbation, marked by strong feeling, and usually an

impulse towards a definite form of behaviour. If the emotion is intense

there is some disturbance of the intellectual functions, a measure of

dissociation, and a tendency towards action of an ungraded or

protopathic character. Beyond this description anything else would

mean an entrance into the controversial field.”

(Lazarus, 1991, p. 36)

“...emotion is not felt experience alone, nor a pattern of neural firing,

nor an action such as smiling. Emotion is the process that emerges

from the dynamic interaction among these components as they occur

in relation to changes in the social and physical context.”

Fogel et al., (1992, p. 129)

Appendices 259

“We define emotions as episodic, relatively short-term, biologically

based patterns of perception, experience, physiology, action, and

communication that occur in response to specific physical and social

challenges and opportunities.”

(Keltner & Gross, 1999, p. 468)

“What is an emotion? My definition of emotion as a phenomenon is

that it is an organized psychophysiological reaction to ongoing

relationships with the environment, most often, but not always,

interpersonal or social. This reaction consists of responses from three

levels of analysis – namely, introspective reports of subjective

experience (often referred to as an affect), overt actions or impulses to

act, and physiological changes that make the emotions organismic.”

(Lazarus, 2000, p. 230)

“Thus, emotion could be very generally defined as a process of

reorganization (integration - disintegration) of the organism-

environment system, expressed most clearly in relation to the

realization of the expected behavioural result.”


“Is an emotion a separate psychological process, or can it be simply

ascribed to the activity of some brain area (for example, limbic area,

prefrontal cortex)? What is the relation between emotional and

cognitive processes? Are emotions only those feelings about which we

260 Appendices

can report? What is the relation between consciousness and

emotions?”


"Emotions denote special organizational aspects related to the

achievement of results: negative emotions refer to disorganization

related to failure in this achievement of the result, and positive

emotions to the integration of action after successful achievement of

the result. As the reorganization of the system is a continuous process,

emotions are always present and there is no action without emotions.”

(Jarvilehto, 2000b, p. 43)

Appendices 261

Appendix B – Emotions during learning in sport survey (phase 1)

262 Appendices

Appendices 263

Appendix C – A Preliminary exploration of emotion items in a golf putting task

264 Appendices

A Preliminary exploration of emotion items in a golf putting task

Following the first phase where emotional items were identified, this

exploratory pilot study was devised to determine whether the items from phase 1

were again identified during an actual learning event in a practical sport

environment. This study used a golf putting task as a vehicle to initially assess the

appropriateness of the items identified in the first phase alongside those previously

included in the preliminary stages of the SEQ development (Jones, et al., 2005).

This exploratory study was not intended to necessarily determine exactly which

items should be included in the final emotions tool, instead it provided an initial

indication of whether participants related to items identified in phase 1 during a

practical sport task. Including the physical putting task also provided the opportunity

to observe performance outcomes in unison with the emotions experienced across a

session, similar to the experimental design of Cooke et al. (2011). It was

hypothesised that during the golf putting task trends would emerge revealing patterns

between putting performance scores and the number/valence of emotions selected.

Furthermore, it was expected that items from the phase 1 list would be selected more

frequently than those from the preliminary SEQ due to the learning context of the

putting task.

Methods

Participants

Thirty-three male (n=24) and female (n=9) undergraduate exercise science

students participated in the pilot investigation (mean age 22.45 ± 4.93 years).

Students were enrolled in an introductory sport and exercise psychology subject and

were a mix of left and right handed putting styles. No students reported having a

golf handicap, with self-reported golf experience ranging from 0 to a maximum of 4

Appendices 265

years. All students provided informed consent to participate in the study which was

also to be used as an assessment piece. Students involved in this study did not

participate any of the phases of the emotion tool development.

Procedure

The putting task was completed during scheduled class time with collected data

also intended for use as a class project. Pairs of participants assembled around

practice holes positioned on a flat artificial grass putting green at an outdoor golf

facility. Using a tape measure, each pair marked a point 3m from the hole to indicate

where the ball was to be placed for each putt. Before each set of putts the

participants were asked to read and follow a set of generic putting instructions that

described five key aspects of a putting stroke (see page 266). Participants completed

five sets of ten putts each from the same position alternatively. Putting attempts

were scored by measuring the distance from the nearest edge of the hole to the centre

of the ball. If the putt was holed the score was a zero.

Prior to each set participants were provided with a list from which they were

asked to identify up to ten emotional items from a randomised list, reflecting what

they were feeling at that moment in relation to the task. All putting scores and

identified emotion items were recorded by the participants on a paper recording

sheet. The list of 54 items included the 39 items from the preliminary SEQ of Jones

et al. (2005), along with 15 unique items carried forward from phase 1 (see Table

4.1). For the purposes of this pilot study, the yet to be confirmed item ‘achievement’

was not included in order for the primary focus to be on contrasting the other 15

confirmed items with those of the preliminary SEQ (see phase 2 for further

discussion of ‘achievement’).

266 Appendices

Putting instructions

Read these instructions through carefully a couple of times. Try to follow them

as closely as possible each time you putt a ball.

In a good putting stroke, the putter should move back and through

smoothly; with the putter very low to the ground.

The body should be bent over until the eyeline is directly over the ball-

to-target line.

The person should feel that a balanced stillness can be maintained in the

body as the arms and putter make the stroke.

Ball position is also critical, as the ideal point of contact is when the

putter is travelling at the lowest point of its swinging arc.

The actual swing of the putter is made by the arms with the hands

serving as the connecting link.

Analysis

Putting scores and identified emotion items were collated into a spreadsheet by

the researcher. Totals, means, and standard deviations for putting scores were

calculated in order to initially inspect the results. Overall putting scores were

subjected to a one-way repeated measures ANOVA with pairwise comparisons

(alpha level < .05), and Bonferroni corrections to control for Type I error. Male and

female scores per set were also compared using a one-way independent ANOVA.

Counts of emotion items identified per set were collated to provide an initial

understanding of the valence (i.e. positive / negative feelings) and number of

emotions associated with each putting set. These results were analysed using one-

Appendices 267

way repeated measure ANOVAs with pairwise comparisons. Bonferroni corrections

were again used to control Type I error and the Greenhouse-Geisser method used to

correct for violation to the sphericity assumption (Field & Hole, 2003). Effect sizes

were also calculated to accompany all ANOVAs using omega (ω), with the following

classifications: small = 0.10; medium = 0.30; large = 0.50) (Field, 2013). As a

measure for determining the change in putting scores and emotion, item selection

between sets’ z scores were calculated. The z scores provided data specifying the

change in putting score, number of positive items selected, and number of negative

items selected through the five putting sets. Changes in z scores were then subjected

to a correlation (Pearson’s correlation coefficient, r) analysis to determine any

relationship between putting score and emotion item values (alpha < .05, 95%

confidence intervals).

Hierarchical cluster analysis using Ward’s method (Ward, 1963) was

performed on the 54 emotion items to establish whether certain items appeared /

clustered together across the sets of trials. This was followed by a one-way ANOVA

to determine whether any cluster was more predominant on any of the five occasions

where emotions were selected by participants. Finally, a second cluster analysis was

performed to determine whether groups of participants could be grouped together in

terms of their putting performance and emotion item (positive and negative) selection

for each set. All statistical analysis was performed using SPSS Version 22.0 (IBM

SPSS Statistics for Windows, 2013).


Putting scores

Results for the putting scores revealed a trend of scores decreasing through the

five sets. Table A1 displays the mean and standard deviations for participants

268 Appendices

overall, along with male and female participant data for each set of trials. Analysis

of overall putting scores for each set revealed a significant main effect between sets

(F 4, 128 = 4.67, P < .05, ω =.32). Pairwise comparisons returned a significant

difference in putting score between sets 1 - 2 (set 1: 1.31 ± 0.70 m; set 2: 1.07 ± 0.67

m; p < .05); 1 – 4 (set 1: 1.31 ± 0.70 m; set 4: 1.03 ± 0.66 m; p < .05); and 1 – 5 (set

1: 1.31 ± 0.70 m; set 5: 1.03 ± 0.83 m; p < .05). Analysis of standard deviations

(variability in putting performance) returned significant main effects across sets

(F3.76, 120.20 = 3.81, p < .05, ω =.14) and a significant pairwise comparison between set

1 and 5 (set 1: 0.93 ± 0.47 m; set 5: 0.71 ± 0.45 m; p < .05). Gender results show

that as a group, female participants missed the hole by significantly greater distances

in comparison to the male participants in set 1 (F1, 31 = 5.94, p < .05, ω = .36); set 2

(F1, 31 = 4.49, p < .05, ω = .31); set 5 (F1, 31 = 7.28, p < .05, ω = .40). Scores were not

significantly different between males and females for set 3 (F1, 31 = 3.79, p > .05, ω =

.31) and set 4 (F1, 31 = 3.04, p > .05, ω = .28).

Appendices 269

Table A1. Summary of putting scores and emotion items for each set.

Putting Scores (m) Emotion Items

Set Overall

Score Mean

± SD

Mean

Variability ±

SD

Male

Mean ± SD

Female

Mean ± SD

Positive

Items

Negative

Items

Total

Items

%

Positive

Mean

Positive ±

SD

Mean

Negative ±

SD

1 1.31 ± 0.70 0.93 ± 0.47 1.14 ± 0.60 1.76 ± 0.79 140 98 238 58.82 4.24 ± 3.17 2.97 ± 2.46

2 1.07 ± 0.67 0.80 ± 0.45 0.92 ± 0.53 1.45 ± 0.87 171 64 235 72.77 5.18 ± 3.62 1.94 ± 2.56

3 1.05 ± 0.70 0.88 ± 0.51 0.91 ± 0.68 1.42 ± 0.67 177 69 245 71.95 5.36 ± 2.91 2.09 ± 1.88

4 1.03 ± 0.66 0.86 ± 0.52 0.91 ± 0.61 1.34 ± 0.72 182 55 237 76.79 5.52 ± 3.14 1.67 ± 1.93

5 1.03 ± 0.83 0.71 ± 0.45 0.81 ± 0.66 1.61 ± 1.00 158 49 207 76.33 4.79 ± 3.30 1.48 ± 1.80

270 Appendices

Emotion items

Detailed counts of emotion items identified during the putting task are

presented in Table A2. Raw data for each item is presented for each set, along with

totals, and the origin of each item (see table caption). These raw results are also

summarised in Table A1 revealing a clear dominance of positive over negative items.

Statistical analysis of positive versus negative emotions per set revealed a significant

main effect (F3.72, 119.15 = 13.88, p < .001, ω = .53). Pairwise comparisons confirmed

that positive (Pos) items were significantly more prominent than negative (Neg)

items for: Set 3 (Pos: 5.36 ± 2.91; Neg: 2.09 ± 1.88; p < .05), Set 4 (Pos: 5.52 ± 3.14;

Neg: 1.67 ± 1.93; p < .05), and Set 5 (Pos: 4.79 ± 3.30; Neg: 1.48 ± 1.80; p < .05).

Pairwise comparisons did not reveal any significant differences across sets when

assessing positive or negative item data separately (e.g. positive items in set 1 versus

positive items in set 2 etc.). Total counts of emotion items (positive combined with

negative) were also not statistically different between sets (F4, 212 = 0.69, p > .05, ω =

.08).

Appendices 271

Table A2. Count of emotion items selected for each set and respective totals presented in descending

order. Origin column indicates where the item originated: SEQ - exclusively from the preliminary list

of 39 items of Jones et al. (2005); Phase 1 - exclusively from the list of unique items in phase 1; Both

– from both the preliminary SEQ and phase 1.

Item Set

1

Set

2

Set

3

Set

4

Set

5

Total Origin Cluster

Competitive 9 12 16 15 12 64 Phase 1 3

Motivated 13 13 8 13 15 62 Both 3

Determined 10 10 20 7 13 60 Phase 1 3

Enjoyment 8 12 15 8 8 51 Phase 1 3

Fun 7 12 14 5 4 42 Phase 1 3

Successful 5 10 7 6 14 42 Phase 1 3

Excited 5 7 14 11 4 41 Both 3

Disappointed 12 5 9 7 8 41 Both 2

Comfortable 8 10 6 7 6 37 SEQ 2

Enthusiastic 12 7 4 8 5 36 SEQ 2

Frustrated 10 3 9 5 9 36 Both 2

Annoyed 13 4 8 8 3 36 Both 2

Happy 4 9 6 9 7 35 Both 2

Energetic 6 15 1 5 4 31 Both 2

Satisfied 4 2 2 12 9 29 Both 2

Irritated 7 7 5 6 3 28 SEQ 2

Cheerful 10 3 5 5 4 27 SEQ 2

Included 7 4 3 9 4 27 Phase 1 2

Eager 4 6 2 9 5 26 Phase 1 2

Pleasure 1 7 2 11 4 25 SEQ 2

Pleased 2 4 8 3 8 25 SEQ 1

Accomplishment 3 3 4 9 5 24 Phase 1 2

Proud 3 5 4 5 3 20 Phase 1 1

Pressured 4 4 8 2 1 19 Both 1

Unhappy 8 3 2 5 1 19 SEQ 1

Confusion 7 3 2 3 2 17 Phase 1 1

Bored 6 2 2 2 4 16 Phase 1 1

Relieved 1 3 3 1 8 16 Phase 1 1

Joy 3 3 4 2 3 15 Both 1

Attacking 3 2 3 4 3 15 SEQ 1

Nervous 6 2 3 2 1 14 Both 1

Alert 2 2 7 1 2 14 SEQ 1

Content 4 4 1 3 2 14 Both 1

Charged 3 0 6 4 0 13 SEQ 1

Fulfilled 1 3 3 4 2 13 Both 1

272 Appendices

Stressed 2 3 4 2 2 13 Both 1

Anxious 2 1 3 2 3 11 Both 1

Uneasy 3 3 3 1 1 11 SEQ 1

Embarrassed 2 2 4 1 1 10 Phase 1 1

Exhilarated 0 5 1 2 2 10 SEQ 1

Tense 4 1 1 2 1 9 SEQ 1

Concerned 3 2 0 2 1 8 SEQ 1

Angry 3 3 1 0 1 8 Both 1

Daring 1 1 3 2 1 8 SEQ 1

Upset 0 2 2 1 3 8 SEQ 1

Apprehensive 0 1 1 1 2 5 SEQ 1

Furious 1 2 0 2 0 5 SEQ 1

Depressed 1 2 1 0 0 4 SEQ 1

Elated 1 1 2 0 0 4 Phase 1 1

Fear 2 0 0 2 0 4 Phase 1 1

Dejected 1 0 1 1 1 4 SEQ 1

Provoked 1 1 1 0 0 3 SEQ 1

Hatred 0 1 0 0 1 2 SEQ 1

Sad 0 1 0 1 0 2 Both 1

Appendices 273

Combined results

From these results it can be observed that (overall) putting performance

improved during the five ‘learning’ sets (significantly from: 1-2, 1- 4, and 1-5 - see

Figure A1). The SD / variation in putting score showed a similar decreasing trend

overall (significantly between sets 1 and 5) However, variability in sets 3 and 4 went

somewhat against this trend. Positive items tended to outnumber negative items at

each stage (significantly for sets 3-5), which is reflected in the percentages of

positive items per set data presented in Table A1. Positive item data showed a trend

of increasing from set 1 through to set 4, and then decreasing from set 4 to 5.

Negative item data somewhat mirrored the positive item data, particularly in the

early stages of the session. For the fifth set, mean putting score appeared to stabilise

and variability was at the lowest value for the session (see Figure A1). Similarly, in

terms of emotion items, the number of negative items identified reached the lowest

value for the session at set 5. Interestingly the count of positive items also decreased

at the final putting set to a value between that of sets 1 and 2.

Therefore, it can be interpreted that at the time of the first set the difference

between positive and negative items was quite low, potentially reflecting the

uncertainty and initial challenges of the task. As the putting session continued the

difference between positive and negative items selected widened, suggesting that the

participants were becoming more comfortable and successful with the task, which is

reflected by the gradual decrease and stabilisation in mean putting scores (and to a

lesser extent variation). The decrease in both positive and negative items at the final

putting set might reflect that participants were now becoming accustomed and

familiar with the task and no longer experiencing the range and intensity of emotions

evident in earlier sets.

274 Appendices

Figure A1. Representation of trends between emotion items and putting scores across sets.

Appendices 275

Correlations of change scores

Correlations for change in z score results revealed no significant correlations

for between sets (e.g. set 1-2, set 3-4) or across the entire session (set 1-5). Change

in putting score and positive item selection returned negative correlations for all sets

and the session as a whole (Table A3). Correlations were generally quite small with

the possible exception of set 3-4. Overall, these results revealed a trend of improved

putting performance (decrease in z score) relating to increased selection of positive

items (increase in z score) and vice versa. The correlation coefficients for changes in

negative item selection in relation to putting scores were again quite small. Data

comparing sets 3-4 and 4-5 did reveal trends of increased selection of negative items

(increase in z scores) relating to a deterioration putting performance (increase in z

scores).

Table A3. Pearson (r) correlation coefficients [95% confidence intervals] for changes in z scores

between putting performance and emotion item selection. S – putting score; P – positive items; N –

negative items. No relationships were statistically significant at the .05 level.

S1-S2 S2-S3 S3-S4 S4-S5 S1-S5

P1-P2 -.102

[-.487, .287]

P2-P3 -.066

[-.358, .235]

P3-P4 -.329

[-.614, .013]

P4-P5 -.036

[-.390, .352]

P1-P5 -.270

[-.571, .075]

N1-N2 .027

[-.241, .329]

N2-N3 -.045

[-.398, .330]

N3-N4 .222

[-.054, .513]

N4-N5 .272

[-.003, .161]

N1-N5 -.065

[-.403, .320]

276 Appendices

Figures A2 and A3 present the relationships between each of the sets

graphically for positive and negative items respectively. Correlations of change

between set 1-2 and 2-3 show little evidence of any trends for either positive or

negative item data. This suggests that during these early sets the participants were

still exploring the task and therefore selected a range of positive and negatively

valenced emotion items, which is supported by the mean and standard deviation

values reported in Table A1. Figures A2a and A3a show that many of the

participants improved their putting scores between sets 1 and 2 with a majority of

cases lying to the left of the y-axis, indicating decreases in putting score (i.e. distance

to hole). Data from set 3 to set 4 (Figures A2c and A3c) revealed the most evident

differences between positive and negative item relationships with putting

performance. While the correlation values remained moderate at best, some trends

can be observed. For positive item data, decreases in putting score related to

increased positive emotion item selection (cases in the top-left quadrant), and

increases in putting score related to decreased item selection (cases in the bottom-

right quadrant). Conversely, negative item data revealed that decreases in putting

score related to decreased negative emotion item selection (bottom-left quadrant),

and increases in putting score corresponded to increased negative item selection (top-

right quadrant). These results suggest that during these sets many participants were

beginning to associate their putting performance (successful or unsuccessful) with

specific emotion items. Similar patterns are evident between sets 4 and 5 however

with a weaker/stronger correlation for positive/negative item data respectively.

Appendices 277

Figure A2. Correlations of changes in z scores for putting performance and number of positive item

selections between sets.

278 Appendices

Figure A3. Correlations of changes in z scores for putting performance and number of negative item

selections between sets.

Appendices 279

Correlation data for change in z scores from the start to the end of the session

(set 1-5) were also small – medium at best. Positive item data revealed a negative

trend indicating that the majority of participants improved their putting performance

throughout the session along with increased selections of positive items (Figure

A4a). This supports the previous findings of putting performance (distance to hole)

improving significantly throughout the session, and the significant predominance of

positive items in sets 3, 4, and 5 (Table A1). Changes in negative emotion item data

from the start to the end of the session were not as defined, this suggests increased

individual variability in the prevalence of negatively oriented items throughout the

session. As expected participant putting scores generally improved, however the

changes in negative item selection were more ambiguous. This could reflect the

predominance of positive items throughout the session and the possibility of

participants continuing to select positive items even when their putting performance

was not improving (i.e. a positive experience of the task regardless of result given

that participants were not golfers and the task was carried out in a University setting).

280 Appendices

Figure A4. Correlations of changes in z scores for putting performance and number of (a) positive

and (b) negative item selections across the session(Sets 1-5)

Appendices 281

Cluster analysis:

Emotion items

Hierarchical cluster analysis of the emotion items using Ward’s method

produced three clusters which contained items that were identified at significantly

different frequencies on each of the five putting sets (see Table A4). Cluster 1

comprised of 33 items, with 16 (48.48%) originating from the phase 1 list. Cluster 2

comprised of 14 items, of which 9 (64.29%) originated from the phase 1 list. All

seven items that grouped to cluster 3 originated from the phase 1 list. Significant

main effects were found between clusters for each putting set: Set 1 (F2, 51 = 23.45, p

< .05, ω =.54); Set 2 (F2, 51 = 51.80, p < .05, ω =.70); Set 3 (F2, 51 = 47.35, p < .05, ω

= .68); Set 4 (F2, 51 = 66.14, p < .05, ω =.74); Set 5 (F2, 51 = 36.52, p < .05, ω =.63).

Cluster 3 items were found to be more frequently selected than cluster 1 items in all

sets; along with cluster 2 items in sets 2, 3, and 5 (see Table A4 for mean and SD

values). Further, cluster 2 items were more frequently selected than cluster 1 items

in all sets. Returning to Table A2, it can be observed that of 21 of the 22 most

selected items overall are from cluster 2, and 3 with the top 7 items making up

cluster 3. Additionally, the 8 most selected items are all included in the list from

phase 1. In contrast, the 32 least selected items overall are from cluster 1 with 16

originating from the preliminary SEQ only. This leaves a further 9 items shared

between the SEQ and the phase 1 list, and 7 items originating exclusively from phase

1.

282 Appendices

Table A4. Items grouping to the three clusters and the origin of these items (Phase 1 list or preliminary SEQ). Mean values display the predominance of Cluster 3 items for

all occasions where emotion items were selected. Significant differences between clusters indicated for: cluster 1 and 2 a

; cluster 2 and 3 b

; cluster 1 and 3 c.

Cluster 1 Cluster 2 Cluster 3

Phase 1

Items

Angry Anxious Bored Accomplished Annoyed Disappointed Competitive Determined Enjoyment

Confusion Content Elated Eager Energetic Frustrated Excited Fun Motivated

Embarrassed Fear Fulfilled Happy Included Satisfied Successful

Joy Nervous Pressure

Proud Relieved Sad

Stressed

SEQ

Items

Alert Apprehensive Attacking Cheerful Comfortable Enthusiastic

Charged Concerned Daring Irritated Pleasure

Dejected Depressed Exhilarated

Furious Hatred Pleased

Provoked Tense Uneasy

Unhappy Upset

Cluster 1 Items Mean± SD Cluster 2 Items Mean ± SD Cluster 3 Items Mean ± SD

Set 1 a c

2.42 ± 2.05 7.21 ± 3.77 8.14 ± 2.85

Set 2 a b c

2.18 ± 1.33 6.07 ± 3.52 10.86 ± 2.04

Set 3 a b

c 2.55 ± 2.20 4.71 ± 2.64 13.43 ± 4.54

Set 4 a c

1.91 ± 1.38 7.86 ± 2.18 9.29 ± 3.77

Set 5 a b c

1.82 ± 1.93 5.43 ± 2.06 10.00 ± 4.65

Appendices 283

Participants

Cluster analysis of individual participant results produced two distinct clusters

based on putting scores and emotion items for each set (see Table A5). Participants

grouped to cluster 1 were found to select significantly more positive items in all five

putting sets than those from cluster 2 (P1-P5). Cluster 1 participants also selected

significantly fewer negative items in the second putting set (N2). Despite no

statistically significant differences, cluster 1 participants also showed a trend of

achieving lower mean putting scores (i.e. closer to the hole) than cluster 2

participants in all sets (S1-S5). These results echo the correlation data with some

participants (cluster 1) showing improved putting performance with associated

increases in positive emotion item selection. On the other hand the remaining

participants (cluster 2) appeared to select fewer positive emotion items, displayed

higher mean putting scores, and on occasion higher counts of negative items.

284 Appendices

Table A5. Comparison of participant gender breakdown, putting scores (S), positive items (P), and

negative items (N) between the two clusters for the five putting sets. *Significant differences at the .05

level.

Cluster 1

n = 17

Cluster 2

n = 16 F (df)

Effect size

(ω)

Gender M: 13

F: 4

M: 11

F: 5 .23

(1, 31) .12

S1 1.24 ± 0.44 1.31 ± 0.48 1.43 (1, 31)

.09

P1 * 5.94 ± 2.88 2.44 ± 2.42 14.20 (1, 31)

.44

N1 2.47 ± 1.84 3.50 ± 2.94 1.47 (1, 31)

.09

S2 1.00 ± 0.56 1.13 ± 0.79 .30 (1, 31)

.11

P2 * 8.24 ± 1.86 1.94 ± 1.57 110.16 (1, 31)

.82

N2 * 0.59 ± 1.18 3.38 ± 2.87 13.60 (1, 31)

.44

S3 0.96 ± 0.76 1.15 ± 0.65 .59 (1, 31)

.09

P3 * 6.88 ± 2.18 3.75 ± 2.77 13.14 (1, 31)

.43

N3 2.24 ± 2.05 1.94 ± 1.73 .20 (1, 31)

.12

S4 0.96 ± 0.64 1.11 ± 0.69 .41 (1, 31)

.10

P4 * 7.59 ± 2.27 3.31 ± 2.36 28.22 (1, 31)

.58

N4 1.53 ± 1.77 1.81 ± 2.14 .17 (1, 31)

.12

S5 0.84 ± 0.74 1.22 ± 0.90 .19 (1, 31)

.12

P5 * 6.29 ± 3.06 3.19 ± 2.81 9.20 (1, 31)

.36

N5 1.18 ± 1.51 1.81 ± 2.07 1.02 (1, 31)

.02

Appendices 285

When considered as a whole, the cluster analysis results indicated that the

emotion items identified in phase 1 were found to be preferred by participants

completing the putting task than those exclusively from the preliminary SEQ of

Jones et al. (2005). To be specific, 16 of the 21 words from clusters 2 and 3 were

from the phase 1 list (see Table A4). A likely reason for this finding is that the

words from phase 1 were originally identified with the context of ‘learning’ in mind

as opposed to ‘competition’ in the case of the SEQ. The predominance of cluster 3,

and to a lesser extent, cluster 2 items across the entire putting session indicates that

these words were most frequently selected as relating to learning. However, the

majority of these words are also positively valenced which, as reported earlier, were

selected significantly more across all sets. Previous work has suggested that negative

emotions might be experienced less frequently, but at higher intensities (Diener, et

al., 1985; Lane & Terry, 2000; Woodman, et al., 2009). Therefore, these results are

not conclusive in terms of identifying the most relevant emotion items for learning,

because the intensity of items (particularly negative items) is somewhat overlooked

when analysing purely on frequency (Larsen & Diener, 1987). Nevertheless, the

finding that the seven most identified items were all from the phase 1 list provides

strong support for the need to develop a emotion questionnaire specific to learning

contexts.

Cluster analysis results concerning putting performance and emotion item

selection for individual participants support the hypothesised link between emotion

and performance (action, outcome) in a practical task. The finding that one group or

cluster of participants selected significantly more positive items in all sets, and also

tended to be more successful with the putting task provides initial evidence for

considering these behavioural aspects in unison. From an alternate perspective, the

286 Appendices

second group of participants generally exhibited higher counts of negative items

alongside their less successful putting performance, reflecting the individualised

nature of learning experiences (Larsen & Diener, 1987). The less pronounced

difference in negative items between clusters again exemplifies how these items

might be identified less frequently, but potentially representing higher intensities

(Diener, et al., 1985).

Considering all the previous findings, the conclusions from this exploratory

pilot study suggest that the majority of items included on the SEQ are not appropriate

for learning contexts. While some items were common between the SEQ and the

phase 1 list, the items exclusive to the SEQ were on the whole neglected, implying

that these items were not considered relevant. Therefore, the development of an

emotion tool specific to learning in sport is warranted. Furthermore, this golf putting

study reveals how performance outcomes (distance to hole), and emotions can

interact during a learning task. Variability and fluctuations in key measures such as

these has the potential to provide insights regarding the holistic experience of a

performer, or group of performers over time.

Appendices 287

Appendix D – Emotions during learning in sport survey (phase 2)

288 Appendices

Appendices 289

290 Appendices

Appendix E – Emotions during learning in sport survey (phase 3)

Appendices 291

292 Appendices

Appendices 293

Appendix F – Perception of speed scale

294 Appendices

Appendices 295

Appendix G – Confrontational interview questions and answers

296 Appendices

Over 1

Over 2

Questions Answers

Who do you think won that over, yourself or the bowler

Probably myself

Why is that? Left a lot more, didn’t try anything stupid and got a couple runs of it

What can you tell me about the bowlers skill level that over?

Definitely improved. Executed his bumper, executed his balls outside off stump . Probably that last ball wasn’t idea I don’t think but thought it was a good over

Do you think the bowler had a game plan?

Yeah that one he did

What was it? Bowl a couple outside off, then bump me. Try and make my feet stop moving

So was he successful? Yeah

Did you have a target in mind that over? No not really. Just try to survive like I did last time. And if there’s a bad ball hit it again like I did

Did that change at any point during the over

Not really

And what can you tell me about your game plan next over?

Just try and keep it the same

Questions Answers

So who do you think won that over, yourself or the bowler

Probably him

Why is that? I scored one off the over and played and missed at a couple

What can you tell me about the bowler’s skill level that over?

He probably wasn’t that happy with how he bowled there. He didn’t hit the same spot 6 balls in a row. Tried too much I think


No not really, I don’t think he had one.

Fair enough, did you have a target in mind?

Not really wasn’t really thinking about it, just thinking bat on ball I think.

What can you tell me about your game plan next over?

Probably leave a couple more and not force much. Just if there is a couple of bad balls just hit them.

So are you doing anything differently this over?

Probably not play at wide ones like I did last over

Appendices 297

How do you plan on scoring runs? Through cover. There’s a big gap through there so if I just punch it through there will be a couple runs. Or if he bowls on my pads then through square leg

Over 3

Questions Answers


Me

Why is that? I hit the balls he probably thought were good for runs and got him a bit frustrated then he tried to bowl on my pads to try and get a wicket on the last ball but misexecuted (sic).

What can you tell me about his skill level that over?

I thought he would probably try something else cause they weren’t working bowling out there, maybe change the field or bumper. But it didn’t

What do you think he did right and what do you think he did wrong

He probably tried to stay patient but it wasn’t. He tried to stay patience but just wasn’t executing.

Do you think he had a game plan? A little bit but it wasn’t very successful

What do you think it was? Bowl outside off, same as before. Just bowl outside off and try make me play and miss or nick one.

Did you have a target in mind that over? Nah not really. Just if I felt like I could hit it, I would hit it


Still the same, still try to keep it simple

Do anything different? Nah

How do you plan on scoring runs? Well he’s moved cover out, so try and get it in the gaps or stay patient and wait for the bad ball and hit that

Any particular gaps If he bowls outside off probably cover or square leg

Whys that? Its where I feel comfortable

Over 4

Questions Answers


Probably me.

Why is that? He had a plan, I played against it and it was successful for me

298 Appendices

What did he do right He tried to, he had a 7-2 offside field and bowled out there. And I kept hitting him through the legside which he wasn’t too happy about

What can you tell me about his skill execution that over?

It probably wasn’t, he might have liked to bowl a little wider I think. He wouldn’t have wanted me playing through the legside


Try and bowl outside off with a 7 -2 off side field, get me frustrated and play against the ball on legside but his execution probably wasn’t good enough for that

Did you have a target in mind Nup


He’s probably going to change the field so just have a look at the field change and think about where the best options to score are

If he keeps it the same? Same then

Okay now he’s changed it, what can you tell me about your game plan next over?

Probably be a bit more reserved in my onside play because he got a midwicket in now so I can’t really risk playing across the line now too much.

How do you plan on scoring runs If he bowls a bit wide maybe try dab him a bit through the slips cause there is 1 slip, 1 gully, no third man, so there is a definite 4 ball there

Over 5

Questions Answers


Myself

Why is that? Probably didn’t execute and he was pretty frustrated throughout the whole over about fielding and bowling. So he didn’t really think about it too much I don’t think

What can you tell me about his execution?

He was a bit frustrated so he wasn’t think about execution he was thinking about pace. I think his pace increased throughput the over

What do youy think he did right and what do you think he did wrong

He probably tried to bowl where he wanted to, but it was just not going where he wanted it to because he mentally not there

Appendices 299


I wasn’t too sure about it, I think he just went for a standard one day ring field, try to bowl stump to stump

Was he successful? I think so yeah

Did you have a target in mind No


Stay the same. Keep trying to hit the gaps. If it’s a bad ball, hit it

How do you plan on scoring runs Same as I did last over, hit the gaps. If it’s on the pads flick it through legside

Any gaps in particular? Square leg, hit that one a bit straight. Probably could have hit it squarer

Over 6

Questions Answers


I think that was even

Why is that? He bowled pretty tightly and I didn’t score too many runs. But I still felt pretty confident

What can you tell me about the bowler’s execution that over?

It improved that over. He wasn’t as frustrated I don’t think

What do you think he did right? Just bowled tight didn’t give me much room to play with

Do much wrong? No I don’t think


Yeah just bowl stump to stump. Make me hit in the air or change what I need to do

Was he successful? I didn’t change what I needed to do. I stayed pretty compact in what my plan was

Did you have a target in mind? No


Going to stay the same. If there is a bad ball, hit it. Try and get a single of it.

Where are you looking to score runs? Again through square leg and then if it’s up try hit it back past him

Over 7

Questions Answers


Me

Why is that? Scored more runs, scored a lot more runs than normal. Just played a couple more aggressive shots - try and get him

300 Appendices

on the back foot.

What can you tell me about his skill execution that over?

I think it was pretty good, kept the same plan as last time and just make me make a mistake, then put a wider one in to tempt me into that wide shot and I fell for that.

So that’s something he did right, did he do anything wrong?

Probably strayed a bit on my pads. I moved across during his delivery stride and he probably saw that and went, “oh I might try bowl on his pads” so I was able to put it away easier than if it was a normal ball

Did you have a target in mind Nah I was a bit more aggressive that over though. Played a couple more big shots than normal

Was there any point during the over where that changed?

Nah I was feeling because he came a bit closer I could be a bit more risky cause he’s a bit quicker and didn’t have to, I just, a bit of power would help me get through it a bit. And there weren’t risky shots per se but like shots I felt comfortable with playing, more aggressive


Probably try and keep it the same that over, score a couple more runs and if the bad ball is there make sure I put it away.

How do you plan on scoring runs Probably couple more fours, not many singles on offer so if I can get it through the field I feel like it’s going to be four Sorry, he’s moved square leg back and taken gully out so there is a big gap behind point. So if he bowls anything wide I’m going to try and hit it through that gap. And probably not as many aerial shots through square leg but still trying to play through there.

Is that what you did last over? No I played a bit more aerial because there wasn’t anyone back I could risk it. But not that someone back I probably don’t want to try getting caught on the fence

Appendices 301

Over 8

Questions Answers


Me

Why is that? Scored a fair few runs

What can you tell me about his skill execution?

Yeah it was pretty good I think. He bowled a decent line and length, but a couple of balls I played an absolutely fluky shot and it went for four. And then just sort of hacked that last ball. So yeah I think he bowled pretty well.

Did he do anything wrong? Nah I don’t think so


Did he? I think so he changed his field a couple of times during that over so he was still thinking which I think he’s just trying to bowl outside off again but a couple of risky shots there from me helped confuse his game plan a bit.

So what do you think he was trying to do?

Bowl outside off and make me play across the leg side again

Was he successful? Yeah I think so yeah

Did you have a target in mind I was probably trying to increase the run rate a little, play a couple of big shots, get him to change the field which was successful

Did your target or game plan change at any point during that over?

A little bit yeah. Sort of just thinking a couple more runs in that over would be nice. So I played that risky shot in that last ball


Probably stay the same, still try to attack him a bit more

How do you plan on scoring runs? Back past him I think and well he’s put that man back at third man so that’s an easy single. Just got to get some bat on it

Over 9

Questions Answers


Him

Why is that? Just bowled really tight, couldn’t do much about it, was trying to invent stuff but it just wasn’t working.

Any reason why? Ah just kept bowling great line and

302 Appendices

length, like yorkers when he needed to, short balls, I couldn’t really move my hands too much.

Yeah, so what did you then think of his skill execution?

His was really good

Did he do anything wrong? Yeah I don’t think so no

Do you think he had a game plan? Yeah he was just trying to bowl either yorkers or short when I was moving around. So it wasn’t, I couldn’t really access the ball too much, get my hands through

So he was pretty successful? Yeah

Did you have a target in mind I was trying to score as many runs as possible

Yeah, how did you go? Not too great.

Did it change at any point during the over?

No - same thing the whole way really


Might go a bit more reserved. Stay a bit more still and try access the ball from there

Why is that? Cause I think it suits his game moving around a lot. Me moving around. So if I just stay still and try wait for the ball to come into my zone, it’ll be easier to hit

How do you plan on scoring runs Try hit over covers and back over his head. Then if it’s on my hip dab it for a couple

Do you consider that more reserved than how you played last over?

Yeah cause I was moving around a lot trying to create something that wasn’t there.

So what does reserved mean to you? Well in this scenario reserved means staying still and just waiting for that bad ball. Whereas I was moving around trying to make a bad ball

Appendices 303

Over 10

Questions Answers


I think I ended up winning it after that last ball. I wasn’t trying to play too many shots, but I suppose when its last ball I might as well. And the lucky four helped my cause, I only got singles off every other ball pretty sure

What can you tell me about the bowler’s skill level that over?

I think I helped it go down because I didn’t move around. He likes when people move around and try sort it out but I don’t think it was as good as his last over

Did he do anything right? Yeah he bowled pretty tight areas, I know that’s his best ball so I try use it to my advantage by moving around in my crease. But not move too much

Did he do anything wrong? Probably bowled a bit too much on my toes and my body the first three or four balls because, that was his, I think that was his plan though. Just cramp it up and then get me singles. But no, cramp me up for room and then he went away from that. Which wasn’t very good

Did you have a target in mind? My goal was not get out, and then get six singles or if it was there, a four or a couple.

And you mentioned you changed on the last ball?

Yeah I was, it was last ball I might as well risk it. And if I get four, I get four, otherwise it was going to be a dot because I wasn’t trying to go in the air I was going to try play that shot and it was just successful

304 Appendices

End of Game

Questions Answers

How do you think you went overall for the 10 overs

I think I did pretty well

In terms of trying to achieve whatever your goal was?

Goal was to not get out and score as many runs as possible. And I think I did that

Do you feel like you were consistent throughout the whole test? Or were there periods where you did better than..

Periods at the start and end where I wasn’t trying to score too many runs. Like mix patches. But I think you go through that in cricket, you go through that in batting

If you were to rate yourself from 1 – 10 on how successful you were?

Probably a 7

What do you think you might have done wrong?

Probably got a bit frustrated with myself towards the end. Once I went back to my plan I felt I was more successful?

Was there any point in particular did you feel you fell off or got frustrated?

Yeah I can’t remember what over it was that over where I was running moving doing everything under the sun. But next ball, next over sorry I felt better

Was there any reason why you changed your game plan and started moving around?

Trying to score a few more runs. But then I realised that’s not my game plan to move around too much and I should instead stay still

AFFECTIVE LEARNING ESIGN A PRINCIPLED APPROACH TO … · Affective Learning Design: A Principled...

Documents

Transcript of AFFECTIVE LEARNING ESIGN A PRINCIPLED APPROACH TO … · Affective Learning Design: A Principled...